{
 "renderer": "transformers 5.23.0",
 "at": "2026-11-14T11:50:17Z",
 "user_mark": "MV2USER hello, what do you remember about me?",
 "ctx_mark": "MV2CTX memory: the user likes short answers",
 "tool_name_placeholder": "recall_context",
 "ids": {
  "call_hex24": "call_0123456789abcdef01234567",
  "uuid4": "3f2b8c1e-9a47-4d2e-b6f1-0c5d7e8a9b10",
  "alnum9": "a1B2c3D4e"
 },
 "rows": [
  {
   "repo": "Qwen/Qwen3-8B",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/Qwen/Qwen3-8B/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "a55ee1b1660128b7",
   "shape": "A_0130_system_prompt",
   "args": "dict",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "user likes short answers\n\n# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>\n{\"type\": \"function\", \"function\": {\"name\": \"recall_context\", \"description\": \"Recall saved memories.\", \"parameters\": {\"type\": \"object\", \"properties\": {\"query\": {\"type\": \"string\"}}, \"required\": []}}}\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n<|im_start|>user\nMV2USER hello, what do you remember about me?<|im_end|>\n<|im_start|>assistant\n"
  },
  {
   "repo": "Qwen/Qwen3-8B",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/Qwen/Qwen3-8B/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "a55ee1b1660128b7",
   "shape": "B_0131_hidden_tool",
   "args": "dict",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "name\": \"recall_context\", \"description\": \"Recall saved memories.\", \"parameters\": {\"type\": \"object\", \"properties\": {\"query\": {\"type\": \"string\"}}, \"required\": []}}}\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n<|im_start|>user\nMV2USER hello, what do you remember about me?<|im_end|>\n<|im_start|>assistant\n<tool_call>\n{\"name\": \"recall_context\", \"arguments\": {\"query\": \"user\"}}\n</tool_call><|im_end|>\n<|im_start|>user\n<tool_response>\nMV2CTX memory: the user likes short answers\n</tool_response><|im_end|>\n<|im_start|>assistant\n"
  },
  {
   "repo": "Qwen/Qwen3-8B",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/Qwen/Qwen3-8B/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "a55ee1b1660128b7",
   "shape": "B_0131_hidden_tool",
   "args": "dict",
   "id": "uuid4",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "name\": \"recall_context\", \"description\": \"Recall saved memories.\", \"parameters\": {\"type\": \"object\", \"properties\": {\"query\": {\"type\": \"string\"}}, \"required\": []}}}\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n<|im_start|>user\nMV2USER hello, what do you remember about me?<|im_end|>\n<|im_start|>assistant\n<tool_call>\n{\"name\": \"recall_context\", \"arguments\": {\"query\": \"user\"}}\n</tool_call><|im_end|>\n<|im_start|>user\n<tool_response>\nMV2CTX memory: the user likes short answers\n</tool_response><|im_end|>\n<|im_start|>assistant\n"
  },
  {
   "repo": "Qwen/Qwen3-8B",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/Qwen/Qwen3-8B/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "a55ee1b1660128b7",
   "shape": "B_0131_hidden_tool",
   "args": "dict",
   "id": "alnum9",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "name\": \"recall_context\", \"description\": \"Recall saved memories.\", \"parameters\": {\"type\": \"object\", \"properties\": {\"query\": {\"type\": \"string\"}}, \"required\": []}}}\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n<|im_start|>user\nMV2USER hello, what do you remember about me?<|im_end|>\n<|im_start|>assistant\n<tool_call>\n{\"name\": \"recall_context\", \"arguments\": {\"query\": \"user\"}}\n</tool_call><|im_end|>\n<|im_start|>user\n<tool_response>\nMV2CTX memory: the user likes short answers\n</tool_response><|im_end|>\n<|im_start|>assistant\n"
  },
  {
   "repo": "Qwen/Qwen3-8B",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/Qwen/Qwen3-8B/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "a55ee1b1660128b7",
   "shape": "B_0131_hidden_tool",
   "args": "str",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "name\": \"recall_context\", \"description\": \"Recall saved memories.\", \"parameters\": {\"type\": \"object\", \"properties\": {\"query\": {\"type\": \"string\"}}, \"required\": []}}}\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n<|im_start|>user\nMV2USER hello, what do you remember about me?<|im_end|>\n<|im_start|>assistant\n<tool_call>\n{\"name\": \"recall_context\", \"arguments\": {\"query\": \"user\"}}\n</tool_call><|im_end|>\n<|im_start|>user\n<tool_response>\nMV2CTX memory: the user likes short answers\n</tool_response><|im_end|>\n<|im_start|>assistant\n"
  },
  {
   "repo": "Qwen/Qwen3-8B",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/Qwen/Qwen3-8B/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "a55ee1b1660128b7",
   "shape": "B_0131_hidden_tool",
   "args": "str",
   "id": "uuid4",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "name\": \"recall_context\", \"description\": \"Recall saved memories.\", \"parameters\": {\"type\": \"object\", \"properties\": {\"query\": {\"type\": \"string\"}}, \"required\": []}}}\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n<|im_start|>user\nMV2USER hello, what do you remember about me?<|im_end|>\n<|im_start|>assistant\n<tool_call>\n{\"name\": \"recall_context\", \"arguments\": {\"query\": \"user\"}}\n</tool_call><|im_end|>\n<|im_start|>user\n<tool_response>\nMV2CTX memory: the user likes short answers\n</tool_response><|im_end|>\n<|im_start|>assistant\n"
  },
  {
   "repo": "Qwen/Qwen3-8B",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/Qwen/Qwen3-8B/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "a55ee1b1660128b7",
   "shape": "B_0131_hidden_tool",
   "args": "str",
   "id": "alnum9",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "name\": \"recall_context\", \"description\": \"Recall saved memories.\", \"parameters\": {\"type\": \"object\", \"properties\": {\"query\": {\"type\": \"string\"}}, \"required\": []}}}\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n<|im_start|>user\nMV2USER hello, what do you remember about me?<|im_end|>\n<|im_start|>assistant\n<tool_call>\n{\"name\": \"recall_context\", \"arguments\": {\"query\": \"user\"}}\n</tool_call><|im_end|>\n<|im_start|>user\n<tool_response>\nMV2CTX memory: the user likes short answers\n</tool_response><|im_end|>\n<|im_start|>assistant\n"
  },
  {
   "repo": "Qwen/Qwen3-8B",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/Qwen/Qwen3-8B/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "a55ee1b1660128b7",
   "shape": "C_0131_no_system",
   "args": "dict",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "name\": \"recall_context\", \"description\": \"Recall saved memories.\", \"parameters\": {\"type\": \"object\", \"properties\": {\"query\": {\"type\": \"string\"}}, \"required\": []}}}\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n<|im_start|>user\nMV2USER hello, what do you remember about me?<|im_end|>\n<|im_start|>assistant\n<tool_call>\n{\"name\": \"recall_context\", \"arguments\": {\"query\": \"user\"}}\n</tool_call><|im_end|>\n<|im_start|>user\n<tool_response>\nMV2CTX memory: the user likes short answers\n</tool_response><|im_end|>\n<|im_start|>assistant\n"
  },
  {
   "repo": "Qwen/Qwen3-8B",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/Qwen/Qwen3-8B/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "a55ee1b1660128b7",
   "shape": "D_0131_content_null",
   "args": "dict",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "name\": \"recall_context\", \"description\": \"Recall saved memories.\", \"parameters\": {\"type\": \"object\", \"properties\": {\"query\": {\"type\": \"string\"}}, \"required\": []}}}\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n<|im_start|>user\nMV2USER hello, what do you remember about me?<|im_end|>\n<|im_start|>assistant\n<tool_call>\n{\"name\": \"recall_context\", \"arguments\": {\"query\": \"user\"}}\n</tool_call><|im_end|>\n<|im_start|>user\n<tool_response>\nMV2CTX memory: the user likes short answers\n</tool_response><|im_end|>\n<|im_start|>assistant\n"
  },
  {
   "repo": "Qwen/Qwen3-8B",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/Qwen/Qwen3-8B/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "a55ee1b1660128b7",
   "shape": "E_0131_empty_user",
   "args": "dict",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": false,
   "ctx_in_prompt": true,
   "tail": ":\n<tools>\n{\"type\": \"function\", \"function\": {\"name\": \"recall_context\", \"description\": \"Recall saved memories.\", \"parameters\": {\"type\": \"object\", \"properties\": {\"query\": {\"type\": \"string\"}}, \"required\": []}}}\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n<|im_start|>user\n<|im_end|>\n<|im_start|>assistant\n<tool_call>\n{\"name\": \"recall_context\", \"arguments\": {\"query\": \"user\"}}\n</tool_call><|im_end|>\n<|im_start|>user\n<tool_response>\nMV2CTX memory: the user likes short answers\n</tool_response><|im_end|>\n<|im_start|>assistant\n"
  },
  {
   "repo": "Qwen/Qwen3-30B-A3B",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/Qwen/Qwen3-30B-A3B/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "a55ee1b1660128b7",
   "shape": "A_0130_system_prompt",
   "args": "dict",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "user likes short answers\n\n# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>\n{\"type\": \"function\", \"function\": {\"name\": \"recall_context\", \"description\": \"Recall saved memories.\", \"parameters\": {\"type\": \"object\", \"properties\": {\"query\": {\"type\": \"string\"}}, \"required\": []}}}\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n<|im_start|>user\nMV2USER hello, what do you remember about me?<|im_end|>\n<|im_start|>assistant\n"
  },
  {
   "repo": "Qwen/Qwen3-30B-A3B",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/Qwen/Qwen3-30B-A3B/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "a55ee1b1660128b7",
   "shape": "B_0131_hidden_tool",
   "args": "dict",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "name\": \"recall_context\", \"description\": \"Recall saved memories.\", \"parameters\": {\"type\": \"object\", \"properties\": {\"query\": {\"type\": \"string\"}}, \"required\": []}}}\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n<|im_start|>user\nMV2USER hello, what do you remember about me?<|im_end|>\n<|im_start|>assistant\n<tool_call>\n{\"name\": \"recall_context\", \"arguments\": {\"query\": \"user\"}}\n</tool_call><|im_end|>\n<|im_start|>user\n<tool_response>\nMV2CTX memory: the user likes short answers\n</tool_response><|im_end|>\n<|im_start|>assistant\n"
  },
  {
   "repo": "Qwen/Qwen3-30B-A3B",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/Qwen/Qwen3-30B-A3B/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "a55ee1b1660128b7",
   "shape": "B_0131_hidden_tool",
   "args": "dict",
   "id": "uuid4",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "name\": \"recall_context\", \"description\": \"Recall saved memories.\", \"parameters\": {\"type\": \"object\", \"properties\": {\"query\": {\"type\": \"string\"}}, \"required\": []}}}\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n<|im_start|>user\nMV2USER hello, what do you remember about me?<|im_end|>\n<|im_start|>assistant\n<tool_call>\n{\"name\": \"recall_context\", \"arguments\": {\"query\": \"user\"}}\n</tool_call><|im_end|>\n<|im_start|>user\n<tool_response>\nMV2CTX memory: the user likes short answers\n</tool_response><|im_end|>\n<|im_start|>assistant\n"
  },
  {
   "repo": "Qwen/Qwen3-30B-A3B",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/Qwen/Qwen3-30B-A3B/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "a55ee1b1660128b7",
   "shape": "B_0131_hidden_tool",
   "args": "dict",
   "id": "alnum9",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "name\": \"recall_context\", \"description\": \"Recall saved memories.\", \"parameters\": {\"type\": \"object\", \"properties\": {\"query\": {\"type\": \"string\"}}, \"required\": []}}}\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n<|im_start|>user\nMV2USER hello, what do you remember about me?<|im_end|>\n<|im_start|>assistant\n<tool_call>\n{\"name\": \"recall_context\", \"arguments\": {\"query\": \"user\"}}\n</tool_call><|im_end|>\n<|im_start|>user\n<tool_response>\nMV2CTX memory: the user likes short answers\n</tool_response><|im_end|>\n<|im_start|>assistant\n"
  },
  {
   "repo": "Qwen/Qwen3-30B-A3B",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/Qwen/Qwen3-30B-A3B/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "a55ee1b1660128b7",
   "shape": "B_0131_hidden_tool",
   "args": "str",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "name\": \"recall_context\", \"description\": \"Recall saved memories.\", \"parameters\": {\"type\": \"object\", \"properties\": {\"query\": {\"type\": \"string\"}}, \"required\": []}}}\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n<|im_start|>user\nMV2USER hello, what do you remember about me?<|im_end|>\n<|im_start|>assistant\n<tool_call>\n{\"name\": \"recall_context\", \"arguments\": {\"query\": \"user\"}}\n</tool_call><|im_end|>\n<|im_start|>user\n<tool_response>\nMV2CTX memory: the user likes short answers\n</tool_response><|im_end|>\n<|im_start|>assistant\n"
  },
  {
   "repo": "Qwen/Qwen3-30B-A3B",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/Qwen/Qwen3-30B-A3B/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "a55ee1b1660128b7",
   "shape": "B_0131_hidden_tool",
   "args": "str",
   "id": "uuid4",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "name\": \"recall_context\", \"description\": \"Recall saved memories.\", \"parameters\": {\"type\": \"object\", \"properties\": {\"query\": {\"type\": \"string\"}}, \"required\": []}}}\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n<|im_start|>user\nMV2USER hello, what do you remember about me?<|im_end|>\n<|im_start|>assistant\n<tool_call>\n{\"name\": \"recall_context\", \"arguments\": {\"query\": \"user\"}}\n</tool_call><|im_end|>\n<|im_start|>user\n<tool_response>\nMV2CTX memory: the user likes short answers\n</tool_response><|im_end|>\n<|im_start|>assistant\n"
  },
  {
   "repo": "Qwen/Qwen3-30B-A3B",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/Qwen/Qwen3-30B-A3B/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "a55ee1b1660128b7",
   "shape": "B_0131_hidden_tool",
   "args": "str",
   "id": "alnum9",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "name\": \"recall_context\", \"description\": \"Recall saved memories.\", \"parameters\": {\"type\": \"object\", \"properties\": {\"query\": {\"type\": \"string\"}}, \"required\": []}}}\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n<|im_start|>user\nMV2USER hello, what do you remember about me?<|im_end|>\n<|im_start|>assistant\n<tool_call>\n{\"name\": \"recall_context\", \"arguments\": {\"query\": \"user\"}}\n</tool_call><|im_end|>\n<|im_start|>user\n<tool_response>\nMV2CTX memory: the user likes short answers\n</tool_response><|im_end|>\n<|im_start|>assistant\n"
  },
  {
   "repo": "Qwen/Qwen3-30B-A3B",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/Qwen/Qwen3-30B-A3B/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "a55ee1b1660128b7",
   "shape": "C_0131_no_system",
   "args": "dict",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "name\": \"recall_context\", \"description\": \"Recall saved memories.\", \"parameters\": {\"type\": \"object\", \"properties\": {\"query\": {\"type\": \"string\"}}, \"required\": []}}}\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n<|im_start|>user\nMV2USER hello, what do you remember about me?<|im_end|>\n<|im_start|>assistant\n<tool_call>\n{\"name\": \"recall_context\", \"arguments\": {\"query\": \"user\"}}\n</tool_call><|im_end|>\n<|im_start|>user\n<tool_response>\nMV2CTX memory: the user likes short answers\n</tool_response><|im_end|>\n<|im_start|>assistant\n"
  },
  {
   "repo": "Qwen/Qwen3-30B-A3B",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/Qwen/Qwen3-30B-A3B/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "a55ee1b1660128b7",
   "shape": "D_0131_content_null",
   "args": "dict",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "name\": \"recall_context\", \"description\": \"Recall saved memories.\", \"parameters\": {\"type\": \"object\", \"properties\": {\"query\": {\"type\": \"string\"}}, \"required\": []}}}\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n<|im_start|>user\nMV2USER hello, what do you remember about me?<|im_end|>\n<|im_start|>assistant\n<tool_call>\n{\"name\": \"recall_context\", \"arguments\": {\"query\": \"user\"}}\n</tool_call><|im_end|>\n<|im_start|>user\n<tool_response>\nMV2CTX memory: the user likes short answers\n</tool_response><|im_end|>\n<|im_start|>assistant\n"
  },
  {
   "repo": "Qwen/Qwen3-30B-A3B",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/Qwen/Qwen3-30B-A3B/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "a55ee1b1660128b7",
   "shape": "E_0131_empty_user",
   "args": "dict",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": false,
   "ctx_in_prompt": true,
   "tail": ":\n<tools>\n{\"type\": \"function\", \"function\": {\"name\": \"recall_context\", \"description\": \"Recall saved memories.\", \"parameters\": {\"type\": \"object\", \"properties\": {\"query\": {\"type\": \"string\"}}, \"required\": []}}}\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n<|im_start|>user\n<|im_end|>\n<|im_start|>assistant\n<tool_call>\n{\"name\": \"recall_context\", \"arguments\": {\"query\": \"user\"}}\n</tool_call><|im_end|>\n<|im_start|>user\n<tool_response>\nMV2CTX memory: the user likes short answers\n</tool_response><|im_end|>\n<|im_start|>assistant\n"
  },
  {
   "repo": "Qwen/Qwen3-4B-Instruct-2507",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/Qwen/Qwen3-4B-Instruct-2507/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "64f85b198065d0fb",
   "shape": "A_0130_system_prompt",
   "args": "dict",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "user likes short answers\n\n# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>\n{\"type\": \"function\", \"function\": {\"name\": \"recall_context\", \"description\": \"Recall saved memories.\", \"parameters\": {\"type\": \"object\", \"properties\": {\"query\": {\"type\": \"string\"}}, \"required\": []}}}\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n<|im_start|>user\nMV2USER hello, what do you remember about me?<|im_end|>\n<|im_start|>assistant\n"
  },
  {
   "repo": "Qwen/Qwen3-4B-Instruct-2507",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/Qwen/Qwen3-4B-Instruct-2507/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "64f85b198065d0fb",
   "shape": "B_0131_hidden_tool",
   "args": "dict",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "name\": \"recall_context\", \"description\": \"Recall saved memories.\", \"parameters\": {\"type\": \"object\", \"properties\": {\"query\": {\"type\": \"string\"}}, \"required\": []}}}\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n<|im_start|>user\nMV2USER hello, what do you remember about me?<|im_end|>\n<|im_start|>assistant\n<tool_call>\n{\"name\": \"recall_context\", \"arguments\": {\"query\": \"user\"}}\n</tool_call><|im_end|>\n<|im_start|>user\n<tool_response>\nMV2CTX memory: the user likes short answers\n</tool_response><|im_end|>\n<|im_start|>assistant\n"
  },
  {
   "repo": "Qwen/Qwen3-4B-Instruct-2507",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/Qwen/Qwen3-4B-Instruct-2507/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "64f85b198065d0fb",
   "shape": "B_0131_hidden_tool",
   "args": "dict",
   "id": "uuid4",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "name\": \"recall_context\", \"description\": \"Recall saved memories.\", \"parameters\": {\"type\": \"object\", \"properties\": {\"query\": {\"type\": \"string\"}}, \"required\": []}}}\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n<|im_start|>user\nMV2USER hello, what do you remember about me?<|im_end|>\n<|im_start|>assistant\n<tool_call>\n{\"name\": \"recall_context\", \"arguments\": {\"query\": \"user\"}}\n</tool_call><|im_end|>\n<|im_start|>user\n<tool_response>\nMV2CTX memory: the user likes short answers\n</tool_response><|im_end|>\n<|im_start|>assistant\n"
  },
  {
   "repo": "Qwen/Qwen3-4B-Instruct-2507",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/Qwen/Qwen3-4B-Instruct-2507/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "64f85b198065d0fb",
   "shape": "B_0131_hidden_tool",
   "args": "dict",
   "id": "alnum9",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "name\": \"recall_context\", \"description\": \"Recall saved memories.\", \"parameters\": {\"type\": \"object\", \"properties\": {\"query\": {\"type\": \"string\"}}, \"required\": []}}}\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n<|im_start|>user\nMV2USER hello, what do you remember about me?<|im_end|>\n<|im_start|>assistant\n<tool_call>\n{\"name\": \"recall_context\", \"arguments\": {\"query\": \"user\"}}\n</tool_call><|im_end|>\n<|im_start|>user\n<tool_response>\nMV2CTX memory: the user likes short answers\n</tool_response><|im_end|>\n<|im_start|>assistant\n"
  },
  {
   "repo": "Qwen/Qwen3-4B-Instruct-2507",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/Qwen/Qwen3-4B-Instruct-2507/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "64f85b198065d0fb",
   "shape": "B_0131_hidden_tool",
   "args": "str",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "name\": \"recall_context\", \"description\": \"Recall saved memories.\", \"parameters\": {\"type\": \"object\", \"properties\": {\"query\": {\"type\": \"string\"}}, \"required\": []}}}\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n<|im_start|>user\nMV2USER hello, what do you remember about me?<|im_end|>\n<|im_start|>assistant\n<tool_call>\n{\"name\": \"recall_context\", \"arguments\": {\"query\": \"user\"}}\n</tool_call><|im_end|>\n<|im_start|>user\n<tool_response>\nMV2CTX memory: the user likes short answers\n</tool_response><|im_end|>\n<|im_start|>assistant\n"
  },
  {
   "repo": "Qwen/Qwen3-4B-Instruct-2507",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/Qwen/Qwen3-4B-Instruct-2507/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "64f85b198065d0fb",
   "shape": "B_0131_hidden_tool",
   "args": "str",
   "id": "uuid4",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "name\": \"recall_context\", \"description\": \"Recall saved memories.\", \"parameters\": {\"type\": \"object\", \"properties\": {\"query\": {\"type\": \"string\"}}, \"required\": []}}}\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n<|im_start|>user\nMV2USER hello, what do you remember about me?<|im_end|>\n<|im_start|>assistant\n<tool_call>\n{\"name\": \"recall_context\", \"arguments\": {\"query\": \"user\"}}\n</tool_call><|im_end|>\n<|im_start|>user\n<tool_response>\nMV2CTX memory: the user likes short answers\n</tool_response><|im_end|>\n<|im_start|>assistant\n"
  },
  {
   "repo": "Qwen/Qwen3-4B-Instruct-2507",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/Qwen/Qwen3-4B-Instruct-2507/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "64f85b198065d0fb",
   "shape": "B_0131_hidden_tool",
   "args": "str",
   "id": "alnum9",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "name\": \"recall_context\", \"description\": \"Recall saved memories.\", \"parameters\": {\"type\": \"object\", \"properties\": {\"query\": {\"type\": \"string\"}}, \"required\": []}}}\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n<|im_start|>user\nMV2USER hello, what do you remember about me?<|im_end|>\n<|im_start|>assistant\n<tool_call>\n{\"name\": \"recall_context\", \"arguments\": {\"query\": \"user\"}}\n</tool_call><|im_end|>\n<|im_start|>user\n<tool_response>\nMV2CTX memory: the user likes short answers\n</tool_response><|im_end|>\n<|im_start|>assistant\n"
  },
  {
   "repo": "Qwen/Qwen3-4B-Instruct-2507",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/Qwen/Qwen3-4B-Instruct-2507/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "64f85b198065d0fb",
   "shape": "C_0131_no_system",
   "args": "dict",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "name\": \"recall_context\", \"description\": \"Recall saved memories.\", \"parameters\": {\"type\": \"object\", \"properties\": {\"query\": {\"type\": \"string\"}}, \"required\": []}}}\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n<|im_start|>user\nMV2USER hello, what do you remember about me?<|im_end|>\n<|im_start|>assistant\n<tool_call>\n{\"name\": \"recall_context\", \"arguments\": {\"query\": \"user\"}}\n</tool_call><|im_end|>\n<|im_start|>user\n<tool_response>\nMV2CTX memory: the user likes short answers\n</tool_response><|im_end|>\n<|im_start|>assistant\n"
  },
  {
   "repo": "Qwen/Qwen3-4B-Instruct-2507",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/Qwen/Qwen3-4B-Instruct-2507/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "64f85b198065d0fb",
   "shape": "D_0131_content_null",
   "args": "dict",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "name\": \"recall_context\", \"description\": \"Recall saved memories.\", \"parameters\": {\"type\": \"object\", \"properties\": {\"query\": {\"type\": \"string\"}}, \"required\": []}}}\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n<|im_start|>user\nMV2USER hello, what do you remember about me?<|im_end|>\n<|im_start|>assistant\n<tool_call>\n{\"name\": \"recall_context\", \"arguments\": {\"query\": \"user\"}}\n</tool_call><|im_end|>\n<|im_start|>user\n<tool_response>\nMV2CTX memory: the user likes short answers\n</tool_response><|im_end|>\n<|im_start|>assistant\n"
  },
  {
   "repo": "Qwen/Qwen3-4B-Instruct-2507",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/Qwen/Qwen3-4B-Instruct-2507/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "64f85b198065d0fb",
   "shape": "E_0131_empty_user",
   "args": "dict",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": false,
   "ctx_in_prompt": true,
   "tail": ":\n<tools>\n{\"type\": \"function\", \"function\": {\"name\": \"recall_context\", \"description\": \"Recall saved memories.\", \"parameters\": {\"type\": \"object\", \"properties\": {\"query\": {\"type\": \"string\"}}, \"required\": []}}}\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n<|im_start|>user\n<|im_end|>\n<|im_start|>assistant\n<tool_call>\n{\"name\": \"recall_context\", \"arguments\": {\"query\": \"user\"}}\n</tool_call><|im_end|>\n<|im_start|>user\n<tool_response>\nMV2CTX memory: the user likes short answers\n</tool_response><|im_end|>\n<|im_start|>assistant\n"
  },
  {
   "repo": "Qwen/Qwen3.5-9B",
   "fetched": true,
   "file": "chat_template.jinja",
   "url": "https://huggingface.co/Qwen/Qwen3.5-9B/resolve/main/chat_template.jinja",
   "template": "default",
   "sha16": "a4aee8afcf2e0711",
   "shape": "A_0130_system_prompt",
   "args": "dict",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "nction>\n</tool_call>\n\n<IMPORTANT>\nReminder:\n- Function calls MUST follow the specified format: an inner <function=...></function> block must be nested within <tool_call></tool_call> XML tags\n- Required parameters MUST be specified\n- You may provide optional reasoning for your function call in natural language BEFORE the function call, but NOT after\n- If there is no function call available, answer the question like normal with your current knowledge and do not tell the user about function calls\n</IMPORTANT>\n\nYou are a helpful assistant.\n\nMV2CTX memory: the user likes short answers<|im_end|>\n<|im_start|>user\nMV2USER hello, what do you remember about me?<|im_end|>\n<|im_start|>assistant\n<think>\n"
  },
  {
   "repo": "Qwen/Qwen3.5-9B",
   "fetched": true,
   "file": "chat_template.jinja",
   "url": "https://huggingface.co/Qwen/Qwen3.5-9B/resolve/main/chat_template.jinja",
   "template": "default",
   "sha16": "a4aee8afcf2e0711",
   "shape": "B_0131_hidden_tool",
   "args": "dict",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "ers MUST be specified\n- You may provide optional reasoning for your function call in natural language BEFORE the function call, but NOT after\n- If there is no function call available, answer the question like normal with your current knowledge and do not tell the user about function calls\n</IMPORTANT>\n\nYou are a helpful assistant.<|im_end|>\n<|im_start|>user\nMV2USER hello, what do you remember about me?<|im_end|>\n<|im_start|>assistant\n<think>\n\n</think>\n\n<tool_call>\n<function=recall_context>\n<parameter=query>\nuser\n</parameter>\n</function>\n</tool_call><|im_end|>\n<|im_start|>user\n<tool_response>\nMV2CTX memory: the user likes short answers\n</tool_response><|im_end|>\n<|im_start|>assistant\n<think>\n"
  },
  {
   "repo": "Qwen/Qwen3.5-9B",
   "fetched": true,
   "file": "chat_template.jinja",
   "url": "https://huggingface.co/Qwen/Qwen3.5-9B/resolve/main/chat_template.jinja",
   "template": "default",
   "sha16": "a4aee8afcf2e0711",
   "shape": "B_0131_hidden_tool",
   "args": "dict",
   "id": "uuid4",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "ers MUST be specified\n- You may provide optional reasoning for your function call in natural language BEFORE the function call, but NOT after\n- If there is no function call available, answer the question like normal with your current knowledge and do not tell the user about function calls\n</IMPORTANT>\n\nYou are a helpful assistant.<|im_end|>\n<|im_start|>user\nMV2USER hello, what do you remember about me?<|im_end|>\n<|im_start|>assistant\n<think>\n\n</think>\n\n<tool_call>\n<function=recall_context>\n<parameter=query>\nuser\n</parameter>\n</function>\n</tool_call><|im_end|>\n<|im_start|>user\n<tool_response>\nMV2CTX memory: the user likes short answers\n</tool_response><|im_end|>\n<|im_start|>assistant\n<think>\n"
  },
  {
   "repo": "Qwen/Qwen3.5-9B",
   "fetched": true,
   "file": "chat_template.jinja",
   "url": "https://huggingface.co/Qwen/Qwen3.5-9B/resolve/main/chat_template.jinja",
   "template": "default",
   "sha16": "a4aee8afcf2e0711",
   "shape": "B_0131_hidden_tool",
   "args": "dict",
   "id": "alnum9",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "ers MUST be specified\n- You may provide optional reasoning for your function call in natural language BEFORE the function call, but NOT after\n- If there is no function call available, answer the question like normal with your current knowledge and do not tell the user about function calls\n</IMPORTANT>\n\nYou are a helpful assistant.<|im_end|>\n<|im_start|>user\nMV2USER hello, what do you remember about me?<|im_end|>\n<|im_start|>assistant\n<think>\n\n</think>\n\n<tool_call>\n<function=recall_context>\n<parameter=query>\nuser\n</parameter>\n</function>\n</tool_call><|im_end|>\n<|im_start|>user\n<tool_response>\nMV2CTX memory: the user likes short answers\n</tool_response><|im_end|>\n<|im_start|>assistant\n<think>\n"
  },
  {
   "repo": "Qwen/Qwen3.5-9B",
   "fetched": true,
   "file": "chat_template.jinja",
   "url": "https://huggingface.co/Qwen/Qwen3.5-9B/resolve/main/chat_template.jinja",
   "template": "default",
   "sha16": "a4aee8afcf2e0711",
   "shape": "B_0131_hidden_tool",
   "args": "str",
   "id": "call_hex24",
   "ok": false,
   "error_type": "TypeError",
   "error": "Can only get item pairs from a mapping."
  },
  {
   "repo": "Qwen/Qwen3.5-9B",
   "fetched": true,
   "file": "chat_template.jinja",
   "url": "https://huggingface.co/Qwen/Qwen3.5-9B/resolve/main/chat_template.jinja",
   "template": "default",
   "sha16": "a4aee8afcf2e0711",
   "shape": "B_0131_hidden_tool",
   "args": "str",
   "id": "uuid4",
   "ok": false,
   "error_type": "TypeError",
   "error": "Can only get item pairs from a mapping."
  },
  {
   "repo": "Qwen/Qwen3.5-9B",
   "fetched": true,
   "file": "chat_template.jinja",
   "url": "https://huggingface.co/Qwen/Qwen3.5-9B/resolve/main/chat_template.jinja",
   "template": "default",
   "sha16": "a4aee8afcf2e0711",
   "shape": "B_0131_hidden_tool",
   "args": "str",
   "id": "alnum9",
   "ok": false,
   "error_type": "TypeError",
   "error": "Can only get item pairs from a mapping."
  },
  {
   "repo": "Qwen/Qwen3.5-9B",
   "fetched": true,
   "file": "chat_template.jinja",
   "url": "https://huggingface.co/Qwen/Qwen3.5-9B/resolve/main/chat_template.jinja",
   "template": "default",
   "sha16": "a4aee8afcf2e0711",
   "shape": "C_0131_no_system",
   "args": "dict",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "l> XML tags\n- Required parameters MUST be specified\n- You may provide optional reasoning for your function call in natural language BEFORE the function call, but NOT after\n- If there is no function call available, answer the question like normal with your current knowledge and do not tell the user about function calls\n</IMPORTANT><|im_end|>\n<|im_start|>user\nMV2USER hello, what do you remember about me?<|im_end|>\n<|im_start|>assistant\n<think>\n\n</think>\n\n<tool_call>\n<function=recall_context>\n<parameter=query>\nuser\n</parameter>\n</function>\n</tool_call><|im_end|>\n<|im_start|>user\n<tool_response>\nMV2CTX memory: the user likes short answers\n</tool_response><|im_end|>\n<|im_start|>assistant\n<think>\n"
  },
  {
   "repo": "Qwen/Qwen3.5-9B",
   "fetched": true,
   "file": "chat_template.jinja",
   "url": "https://huggingface.co/Qwen/Qwen3.5-9B/resolve/main/chat_template.jinja",
   "template": "default",
   "sha16": "a4aee8afcf2e0711",
   "shape": "D_0131_content_null",
   "args": "dict",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "ers MUST be specified\n- You may provide optional reasoning for your function call in natural language BEFORE the function call, but NOT after\n- If there is no function call available, answer the question like normal with your current knowledge and do not tell the user about function calls\n</IMPORTANT>\n\nYou are a helpful assistant.<|im_end|>\n<|im_start|>user\nMV2USER hello, what do you remember about me?<|im_end|>\n<|im_start|>assistant\n<think>\n\n</think>\n\n<tool_call>\n<function=recall_context>\n<parameter=query>\nuser\n</parameter>\n</function>\n</tool_call><|im_end|>\n<|im_start|>user\n<tool_response>\nMV2CTX memory: the user likes short answers\n</tool_response><|im_end|>\n<|im_start|>assistant\n<think>\n"
  },
  {
   "repo": "Qwen/Qwen3.5-9B",
   "fetched": true,
   "file": "chat_template.jinja",
   "url": "https://huggingface.co/Qwen/Qwen3.5-9B/resolve/main/chat_template.jinja",
   "template": "default",
   "sha16": "a4aee8afcf2e0711",
   "shape": "E_0131_empty_user",
   "args": "dict",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": false,
   "ctx_in_prompt": true,
   "tail": "call></tool_call> XML tags\n- Required parameters MUST be specified\n- You may provide optional reasoning for your function call in natural language BEFORE the function call, but NOT after\n- If there is no function call available, answer the question like normal with your current knowledge and do not tell the user about function calls\n</IMPORTANT>\n\nYou are a helpful assistant.<|im_end|>\n<|im_start|>user\n<|im_end|>\n<|im_start|>assistant\n<think>\n\n</think>\n\n<tool_call>\n<function=recall_context>\n<parameter=query>\nuser\n</parameter>\n</function>\n</tool_call><|im_end|>\n<|im_start|>user\n<tool_response>\nMV2CTX memory: the user likes short answers\n</tool_response><|im_end|>\n<|im_start|>assistant\n<think>\n"
  },
  {
   "repo": "Qwen/Qwen3.5-35B-A3B",
   "fetched": true,
   "file": "chat_template.jinja",
   "url": "https://huggingface.co/Qwen/Qwen3.5-35B-A3B/resolve/main/chat_template.jinja",
   "template": "default",
   "sha16": "a4aee8afcf2e0711",
   "shape": "A_0130_system_prompt",
   "args": "dict",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "nction>\n</tool_call>\n\n<IMPORTANT>\nReminder:\n- Function calls MUST follow the specified format: an inner <function=...></function> block must be nested within <tool_call></tool_call> XML tags\n- Required parameters MUST be specified\n- You may provide optional reasoning for your function call in natural language BEFORE the function call, but NOT after\n- If there is no function call available, answer the question like normal with your current knowledge and do not tell the user about function calls\n</IMPORTANT>\n\nYou are a helpful assistant.\n\nMV2CTX memory: the user likes short answers<|im_end|>\n<|im_start|>user\nMV2USER hello, what do you remember about me?<|im_end|>\n<|im_start|>assistant\n<think>\n"
  },
  {
   "repo": "Qwen/Qwen3.5-35B-A3B",
   "fetched": true,
   "file": "chat_template.jinja",
   "url": "https://huggingface.co/Qwen/Qwen3.5-35B-A3B/resolve/main/chat_template.jinja",
   "template": "default",
   "sha16": "a4aee8afcf2e0711",
   "shape": "B_0131_hidden_tool",
   "args": "dict",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "ers MUST be specified\n- You may provide optional reasoning for your function call in natural language BEFORE the function call, but NOT after\n- If there is no function call available, answer the question like normal with your current knowledge and do not tell the user about function calls\n</IMPORTANT>\n\nYou are a helpful assistant.<|im_end|>\n<|im_start|>user\nMV2USER hello, what do you remember about me?<|im_end|>\n<|im_start|>assistant\n<think>\n\n</think>\n\n<tool_call>\n<function=recall_context>\n<parameter=query>\nuser\n</parameter>\n</function>\n</tool_call><|im_end|>\n<|im_start|>user\n<tool_response>\nMV2CTX memory: the user likes short answers\n</tool_response><|im_end|>\n<|im_start|>assistant\n<think>\n"
  },
  {
   "repo": "Qwen/Qwen3.5-35B-A3B",
   "fetched": true,
   "file": "chat_template.jinja",
   "url": "https://huggingface.co/Qwen/Qwen3.5-35B-A3B/resolve/main/chat_template.jinja",
   "template": "default",
   "sha16": "a4aee8afcf2e0711",
   "shape": "B_0131_hidden_tool",
   "args": "dict",
   "id": "uuid4",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "ers MUST be specified\n- You may provide optional reasoning for your function call in natural language BEFORE the function call, but NOT after\n- If there is no function call available, answer the question like normal with your current knowledge and do not tell the user about function calls\n</IMPORTANT>\n\nYou are a helpful assistant.<|im_end|>\n<|im_start|>user\nMV2USER hello, what do you remember about me?<|im_end|>\n<|im_start|>assistant\n<think>\n\n</think>\n\n<tool_call>\n<function=recall_context>\n<parameter=query>\nuser\n</parameter>\n</function>\n</tool_call><|im_end|>\n<|im_start|>user\n<tool_response>\nMV2CTX memory: the user likes short answers\n</tool_response><|im_end|>\n<|im_start|>assistant\n<think>\n"
  },
  {
   "repo": "Qwen/Qwen3.5-35B-A3B",
   "fetched": true,
   "file": "chat_template.jinja",
   "url": "https://huggingface.co/Qwen/Qwen3.5-35B-A3B/resolve/main/chat_template.jinja",
   "template": "default",
   "sha16": "a4aee8afcf2e0711",
   "shape": "B_0131_hidden_tool",
   "args": "dict",
   "id": "alnum9",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "ers MUST be specified\n- You may provide optional reasoning for your function call in natural language BEFORE the function call, but NOT after\n- If there is no function call available, answer the question like normal with your current knowledge and do not tell the user about function calls\n</IMPORTANT>\n\nYou are a helpful assistant.<|im_end|>\n<|im_start|>user\nMV2USER hello, what do you remember about me?<|im_end|>\n<|im_start|>assistant\n<think>\n\n</think>\n\n<tool_call>\n<function=recall_context>\n<parameter=query>\nuser\n</parameter>\n</function>\n</tool_call><|im_end|>\n<|im_start|>user\n<tool_response>\nMV2CTX memory: the user likes short answers\n</tool_response><|im_end|>\n<|im_start|>assistant\n<think>\n"
  },
  {
   "repo": "Qwen/Qwen3.5-35B-A3B",
   "fetched": true,
   "file": "chat_template.jinja",
   "url": "https://huggingface.co/Qwen/Qwen3.5-35B-A3B/resolve/main/chat_template.jinja",
   "template": "default",
   "sha16": "a4aee8afcf2e0711",
   "shape": "B_0131_hidden_tool",
   "args": "str",
   "id": "call_hex24",
   "ok": false,
   "error_type": "TypeError",
   "error": "Can only get item pairs from a mapping."
  },
  {
   "repo": "Qwen/Qwen3.5-35B-A3B",
   "fetched": true,
   "file": "chat_template.jinja",
   "url": "https://huggingface.co/Qwen/Qwen3.5-35B-A3B/resolve/main/chat_template.jinja",
   "template": "default",
   "sha16": "a4aee8afcf2e0711",
   "shape": "B_0131_hidden_tool",
   "args": "str",
   "id": "uuid4",
   "ok": false,
   "error_type": "TypeError",
   "error": "Can only get item pairs from a mapping."
  },
  {
   "repo": "Qwen/Qwen3.5-35B-A3B",
   "fetched": true,
   "file": "chat_template.jinja",
   "url": "https://huggingface.co/Qwen/Qwen3.5-35B-A3B/resolve/main/chat_template.jinja",
   "template": "default",
   "sha16": "a4aee8afcf2e0711",
   "shape": "B_0131_hidden_tool",
   "args": "str",
   "id": "alnum9",
   "ok": false,
   "error_type": "TypeError",
   "error": "Can only get item pairs from a mapping."
  },
  {
   "repo": "Qwen/Qwen3.5-35B-A3B",
   "fetched": true,
   "file": "chat_template.jinja",
   "url": "https://huggingface.co/Qwen/Qwen3.5-35B-A3B/resolve/main/chat_template.jinja",
   "template": "default",
   "sha16": "a4aee8afcf2e0711",
   "shape": "C_0131_no_system",
   "args": "dict",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "l> XML tags\n- Required parameters MUST be specified\n- You may provide optional reasoning for your function call in natural language BEFORE the function call, but NOT after\n- If there is no function call available, answer the question like normal with your current knowledge and do not tell the user about function calls\n</IMPORTANT><|im_end|>\n<|im_start|>user\nMV2USER hello, what do you remember about me?<|im_end|>\n<|im_start|>assistant\n<think>\n\n</think>\n\n<tool_call>\n<function=recall_context>\n<parameter=query>\nuser\n</parameter>\n</function>\n</tool_call><|im_end|>\n<|im_start|>user\n<tool_response>\nMV2CTX memory: the user likes short answers\n</tool_response><|im_end|>\n<|im_start|>assistant\n<think>\n"
  },
  {
   "repo": "Qwen/Qwen3.5-35B-A3B",
   "fetched": true,
   "file": "chat_template.jinja",
   "url": "https://huggingface.co/Qwen/Qwen3.5-35B-A3B/resolve/main/chat_template.jinja",
   "template": "default",
   "sha16": "a4aee8afcf2e0711",
   "shape": "D_0131_content_null",
   "args": "dict",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "ers MUST be specified\n- You may provide optional reasoning for your function call in natural language BEFORE the function call, but NOT after\n- If there is no function call available, answer the question like normal with your current knowledge and do not tell the user about function calls\n</IMPORTANT>\n\nYou are a helpful assistant.<|im_end|>\n<|im_start|>user\nMV2USER hello, what do you remember about me?<|im_end|>\n<|im_start|>assistant\n<think>\n\n</think>\n\n<tool_call>\n<function=recall_context>\n<parameter=query>\nuser\n</parameter>\n</function>\n</tool_call><|im_end|>\n<|im_start|>user\n<tool_response>\nMV2CTX memory: the user likes short answers\n</tool_response><|im_end|>\n<|im_start|>assistant\n<think>\n"
  },
  {
   "repo": "Qwen/Qwen3.5-35B-A3B",
   "fetched": true,
   "file": "chat_template.jinja",
   "url": "https://huggingface.co/Qwen/Qwen3.5-35B-A3B/resolve/main/chat_template.jinja",
   "template": "default",
   "sha16": "a4aee8afcf2e0711",
   "shape": "E_0131_empty_user",
   "args": "dict",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": false,
   "ctx_in_prompt": true,
   "tail": "call></tool_call> XML tags\n- Required parameters MUST be specified\n- You may provide optional reasoning for your function call in natural language BEFORE the function call, but NOT after\n- If there is no function call available, answer the question like normal with your current knowledge and do not tell the user about function calls\n</IMPORTANT>\n\nYou are a helpful assistant.<|im_end|>\n<|im_start|>user\n<|im_end|>\n<|im_start|>assistant\n<think>\n\n</think>\n\n<tool_call>\n<function=recall_context>\n<parameter=query>\nuser\n</parameter>\n</function>\n</tool_call><|im_end|>\n<|im_start|>user\n<tool_response>\nMV2CTX memory: the user likes short answers\n</tool_response><|im_end|>\n<|im_start|>assistant\n<think>\n"
  },
  {
   "repo": "Qwen/Qwen2.5-7B-Instruct",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/Qwen/Qwen2.5-7B-Instruct/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "cd8e9439f0570856",
   "shape": "A_0130_system_prompt",
   "args": "dict",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "user likes short answers\n\n# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>\n{\"type\": \"function\", \"function\": {\"name\": \"recall_context\", \"description\": \"Recall saved memories.\", \"parameters\": {\"type\": \"object\", \"properties\": {\"query\": {\"type\": \"string\"}}, \"required\": []}}}\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n<|im_start|>user\nMV2USER hello, what do you remember about me?<|im_end|>\n<|im_start|>assistant\n"
  },
  {
   "repo": "Qwen/Qwen2.5-7B-Instruct",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/Qwen/Qwen2.5-7B-Instruct/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "cd8e9439f0570856",
   "shape": "B_0131_hidden_tool",
   "args": "dict",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "name\": \"recall_context\", \"description\": \"Recall saved memories.\", \"parameters\": {\"type\": \"object\", \"properties\": {\"query\": {\"type\": \"string\"}}, \"required\": []}}}\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n<|im_start|>user\nMV2USER hello, what do you remember about me?<|im_end|>\n<|im_start|>assistant\n<tool_call>\n{\"name\": \"recall_context\", \"arguments\": {\"query\": \"user\"}}\n</tool_call><|im_end|>\n<|im_start|>user\n<tool_response>\nMV2CTX memory: the user likes short answers\n</tool_response><|im_end|>\n<|im_start|>assistant\n"
  },
  {
   "repo": "Qwen/Qwen2.5-7B-Instruct",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/Qwen/Qwen2.5-7B-Instruct/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "cd8e9439f0570856",
   "shape": "B_0131_hidden_tool",
   "args": "dict",
   "id": "uuid4",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "name\": \"recall_context\", \"description\": \"Recall saved memories.\", \"parameters\": {\"type\": \"object\", \"properties\": {\"query\": {\"type\": \"string\"}}, \"required\": []}}}\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n<|im_start|>user\nMV2USER hello, what do you remember about me?<|im_end|>\n<|im_start|>assistant\n<tool_call>\n{\"name\": \"recall_context\", \"arguments\": {\"query\": \"user\"}}\n</tool_call><|im_end|>\n<|im_start|>user\n<tool_response>\nMV2CTX memory: the user likes short answers\n</tool_response><|im_end|>\n<|im_start|>assistant\n"
  },
  {
   "repo": "Qwen/Qwen2.5-7B-Instruct",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/Qwen/Qwen2.5-7B-Instruct/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "cd8e9439f0570856",
   "shape": "B_0131_hidden_tool",
   "args": "dict",
   "id": "alnum9",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "name\": \"recall_context\", \"description\": \"Recall saved memories.\", \"parameters\": {\"type\": \"object\", \"properties\": {\"query\": {\"type\": \"string\"}}, \"required\": []}}}\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n<|im_start|>user\nMV2USER hello, what do you remember about me?<|im_end|>\n<|im_start|>assistant\n<tool_call>\n{\"name\": \"recall_context\", \"arguments\": {\"query\": \"user\"}}\n</tool_call><|im_end|>\n<|im_start|>user\n<tool_response>\nMV2CTX memory: the user likes short answers\n</tool_response><|im_end|>\n<|im_start|>assistant\n"
  },
  {
   "repo": "Qwen/Qwen2.5-7B-Instruct",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/Qwen/Qwen2.5-7B-Instruct/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "cd8e9439f0570856",
   "shape": "B_0131_hidden_tool",
   "args": "str",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": " \"recall_context\", \"description\": \"Recall saved memories.\", \"parameters\": {\"type\": \"object\", \"properties\": {\"query\": {\"type\": \"string\"}}, \"required\": []}}}\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n<|im_start|>user\nMV2USER hello, what do you remember about me?<|im_end|>\n<|im_start|>assistant\n<tool_call>\n{\"name\": \"recall_context\", \"arguments\": \"{\\\"query\\\": \\\"user\\\"}\"}\n</tool_call><|im_end|>\n<|im_start|>user\n<tool_response>\nMV2CTX memory: the user likes short answers\n</tool_response><|im_end|>\n<|im_start|>assistant\n"
  },
  {
   "repo": "Qwen/Qwen2.5-7B-Instruct",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/Qwen/Qwen2.5-7B-Instruct/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "cd8e9439f0570856",
   "shape": "B_0131_hidden_tool",
   "args": "str",
   "id": "uuid4",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": " \"recall_context\", \"description\": \"Recall saved memories.\", \"parameters\": {\"type\": \"object\", \"properties\": {\"query\": {\"type\": \"string\"}}, \"required\": []}}}\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n<|im_start|>user\nMV2USER hello, what do you remember about me?<|im_end|>\n<|im_start|>assistant\n<tool_call>\n{\"name\": \"recall_context\", \"arguments\": \"{\\\"query\\\": \\\"user\\\"}\"}\n</tool_call><|im_end|>\n<|im_start|>user\n<tool_response>\nMV2CTX memory: the user likes short answers\n</tool_response><|im_end|>\n<|im_start|>assistant\n"
  },
  {
   "repo": "Qwen/Qwen2.5-7B-Instruct",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/Qwen/Qwen2.5-7B-Instruct/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "cd8e9439f0570856",
   "shape": "B_0131_hidden_tool",
   "args": "str",
   "id": "alnum9",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": " \"recall_context\", \"description\": \"Recall saved memories.\", \"parameters\": {\"type\": \"object\", \"properties\": {\"query\": {\"type\": \"string\"}}, \"required\": []}}}\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n<|im_start|>user\nMV2USER hello, what do you remember about me?<|im_end|>\n<|im_start|>assistant\n<tool_call>\n{\"name\": \"recall_context\", \"arguments\": \"{\\\"query\\\": \\\"user\\\"}\"}\n</tool_call><|im_end|>\n<|im_start|>user\n<tool_response>\nMV2CTX memory: the user likes short answers\n</tool_response><|im_end|>\n<|im_start|>assistant\n"
  },
  {
   "repo": "Qwen/Qwen2.5-7B-Instruct",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/Qwen/Qwen2.5-7B-Instruct/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "cd8e9439f0570856",
   "shape": "C_0131_no_system",
   "args": "dict",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "name\": \"recall_context\", \"description\": \"Recall saved memories.\", \"parameters\": {\"type\": \"object\", \"properties\": {\"query\": {\"type\": \"string\"}}, \"required\": []}}}\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n<|im_start|>user\nMV2USER hello, what do you remember about me?<|im_end|>\n<|im_start|>assistant\n<tool_call>\n{\"name\": \"recall_context\", \"arguments\": {\"query\": \"user\"}}\n</tool_call><|im_end|>\n<|im_start|>user\n<tool_response>\nMV2CTX memory: the user likes short answers\n</tool_response><|im_end|>\n<|im_start|>assistant\n"
  },
  {
   "repo": "Qwen/Qwen2.5-7B-Instruct",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/Qwen/Qwen2.5-7B-Instruct/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "cd8e9439f0570856",
   "shape": "D_0131_content_null",
   "args": "dict",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "name\": \"recall_context\", \"description\": \"Recall saved memories.\", \"parameters\": {\"type\": \"object\", \"properties\": {\"query\": {\"type\": \"string\"}}, \"required\": []}}}\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n<|im_start|>user\nMV2USER hello, what do you remember about me?<|im_end|>\n<|im_start|>assistant\n<tool_call>\n{\"name\": \"recall_context\", \"arguments\": {\"query\": \"user\"}}\n</tool_call><|im_end|>\n<|im_start|>user\n<tool_response>\nMV2CTX memory: the user likes short answers\n</tool_response><|im_end|>\n<|im_start|>assistant\n"
  },
  {
   "repo": "Qwen/Qwen2.5-7B-Instruct",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/Qwen/Qwen2.5-7B-Instruct/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "cd8e9439f0570856",
   "shape": "E_0131_empty_user",
   "args": "dict",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": false,
   "ctx_in_prompt": true,
   "tail": ":\n<tools>\n{\"type\": \"function\", \"function\": {\"name\": \"recall_context\", \"description\": \"Recall saved memories.\", \"parameters\": {\"type\": \"object\", \"properties\": {\"query\": {\"type\": \"string\"}}, \"required\": []}}}\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n<|im_start|>user\n<|im_end|>\n<|im_start|>assistant\n<tool_call>\n{\"name\": \"recall_context\", \"arguments\": {\"query\": \"user\"}}\n</tool_call><|im_end|>\n<|im_start|>user\n<tool_response>\nMV2CTX memory: the user likes short answers\n</tool_response><|im_end|>\n<|im_start|>assistant\n"
  },
  {
   "repo": "Qwen/QwQ-32B",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/Qwen/QwQ-32B/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "16488ad69d85833e",
   "shape": "A_0130_system_prompt",
   "args": "dict",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "es short answers\n\n# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>\n{\"type\": \"function\", \"function\": {\"name\": \"recall_context\", \"description\": \"Recall saved memories.\", \"parameters\": {\"type\": \"object\", \"properties\": {\"query\": {\"type\": \"string\"}}, \"required\": []}}}\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n<|im_start|>user\nMV2USER hello, what do you remember about me?<|im_end|>\n<|im_start|>assistant\n<think>\n"
  },
  {
   "repo": "Qwen/QwQ-32B",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/Qwen/QwQ-32B/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "16488ad69d85833e",
   "shape": "B_0131_hidden_tool",
   "args": "dict",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "recall_context\", \"description\": \"Recall saved memories.\", \"parameters\": {\"type\": \"object\", \"properties\": {\"query\": {\"type\": \"string\"}}, \"required\": []}}}\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n<|im_start|>user\nMV2USER hello, what do you remember about me?<|im_end|>\n<|im_start|>assistant\n<tool_call>\n{\"name\": \"recall_context\", \"arguments\": {\"query\": \"user\"}}\n</tool_call><|im_end|>\n<|im_start|>user\n<tool_response>\nMV2CTX memory: the user likes short answers\n</tool_response><|im_end|>\n<|im_start|>assistant\n<think>\n"
  },
  {
   "repo": "Qwen/QwQ-32B",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/Qwen/QwQ-32B/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "16488ad69d85833e",
   "shape": "B_0131_hidden_tool",
   "args": "dict",
   "id": "uuid4",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "recall_context\", \"description\": \"Recall saved memories.\", \"parameters\": {\"type\": \"object\", \"properties\": {\"query\": {\"type\": \"string\"}}, \"required\": []}}}\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n<|im_start|>user\nMV2USER hello, what do you remember about me?<|im_end|>\n<|im_start|>assistant\n<tool_call>\n{\"name\": \"recall_context\", \"arguments\": {\"query\": \"user\"}}\n</tool_call><|im_end|>\n<|im_start|>user\n<tool_response>\nMV2CTX memory: the user likes short answers\n</tool_response><|im_end|>\n<|im_start|>assistant\n<think>\n"
  },
  {
   "repo": "Qwen/QwQ-32B",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/Qwen/QwQ-32B/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "16488ad69d85833e",
   "shape": "B_0131_hidden_tool",
   "args": "dict",
   "id": "alnum9",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "recall_context\", \"description\": \"Recall saved memories.\", \"parameters\": {\"type\": \"object\", \"properties\": {\"query\": {\"type\": \"string\"}}, \"required\": []}}}\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n<|im_start|>user\nMV2USER hello, what do you remember about me?<|im_end|>\n<|im_start|>assistant\n<tool_call>\n{\"name\": \"recall_context\", \"arguments\": {\"query\": \"user\"}}\n</tool_call><|im_end|>\n<|im_start|>user\n<tool_response>\nMV2CTX memory: the user likes short answers\n</tool_response><|im_end|>\n<|im_start|>assistant\n<think>\n"
  },
  {
   "repo": "Qwen/QwQ-32B",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/Qwen/QwQ-32B/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "16488ad69d85833e",
   "shape": "B_0131_hidden_tool",
   "args": "str",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "_context\", \"description\": \"Recall saved memories.\", \"parameters\": {\"type\": \"object\", \"properties\": {\"query\": {\"type\": \"string\"}}, \"required\": []}}}\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n<|im_start|>user\nMV2USER hello, what do you remember about me?<|im_end|>\n<|im_start|>assistant\n<tool_call>\n{\"name\": \"recall_context\", \"arguments\": \"{\\\"query\\\": \\\"user\\\"}\"}\n</tool_call><|im_end|>\n<|im_start|>user\n<tool_response>\nMV2CTX memory: the user likes short answers\n</tool_response><|im_end|>\n<|im_start|>assistant\n<think>\n"
  },
  {
   "repo": "Qwen/QwQ-32B",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/Qwen/QwQ-32B/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "16488ad69d85833e",
   "shape": "B_0131_hidden_tool",
   "args": "str",
   "id": "uuid4",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "_context\", \"description\": \"Recall saved memories.\", \"parameters\": {\"type\": \"object\", \"properties\": {\"query\": {\"type\": \"string\"}}, \"required\": []}}}\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n<|im_start|>user\nMV2USER hello, what do you remember about me?<|im_end|>\n<|im_start|>assistant\n<tool_call>\n{\"name\": \"recall_context\", \"arguments\": \"{\\\"query\\\": \\\"user\\\"}\"}\n</tool_call><|im_end|>\n<|im_start|>user\n<tool_response>\nMV2CTX memory: the user likes short answers\n</tool_response><|im_end|>\n<|im_start|>assistant\n<think>\n"
  },
  {
   "repo": "Qwen/QwQ-32B",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/Qwen/QwQ-32B/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "16488ad69d85833e",
   "shape": "B_0131_hidden_tool",
   "args": "str",
   "id": "alnum9",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "_context\", \"description\": \"Recall saved memories.\", \"parameters\": {\"type\": \"object\", \"properties\": {\"query\": {\"type\": \"string\"}}, \"required\": []}}}\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n<|im_start|>user\nMV2USER hello, what do you remember about me?<|im_end|>\n<|im_start|>assistant\n<tool_call>\n{\"name\": \"recall_context\", \"arguments\": \"{\\\"query\\\": \\\"user\\\"}\"}\n</tool_call><|im_end|>\n<|im_start|>user\n<tool_response>\nMV2CTX memory: the user likes short answers\n</tool_response><|im_end|>\n<|im_start|>assistant\n<think>\n"
  },
  {
   "repo": "Qwen/QwQ-32B",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/Qwen/QwQ-32B/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "16488ad69d85833e",
   "shape": "C_0131_no_system",
   "args": "dict",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "recall_context\", \"description\": \"Recall saved memories.\", \"parameters\": {\"type\": \"object\", \"properties\": {\"query\": {\"type\": \"string\"}}, \"required\": []}}}\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n<|im_start|>user\nMV2USER hello, what do you remember about me?<|im_end|>\n<|im_start|>assistant\n<tool_call>\n{\"name\": \"recall_context\", \"arguments\": {\"query\": \"user\"}}\n</tool_call><|im_end|>\n<|im_start|>user\n<tool_response>\nMV2CTX memory: the user likes short answers\n</tool_response><|im_end|>\n<|im_start|>assistant\n<think>\n"
  },
  {
   "repo": "Qwen/QwQ-32B",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/Qwen/QwQ-32B/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "16488ad69d85833e",
   "shape": "D_0131_content_null",
   "args": "dict",
   "id": "call_hex24",
   "ok": false,
   "error_type": "UndefinedError",
   "error": "'None' has no attribute 'split'"
  },
  {
   "repo": "Qwen/QwQ-32B",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/Qwen/QwQ-32B/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "16488ad69d85833e",
   "shape": "E_0131_empty_user",
   "args": "dict",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": false,
   "ctx_in_prompt": true,
   "tail": ">\n{\"type\": \"function\", \"function\": {\"name\": \"recall_context\", \"description\": \"Recall saved memories.\", \"parameters\": {\"type\": \"object\", \"properties\": {\"query\": {\"type\": \"string\"}}, \"required\": []}}}\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n<|im_start|>user\n<|im_end|>\n<|im_start|>assistant\n<tool_call>\n{\"name\": \"recall_context\", \"arguments\": {\"query\": \"user\"}}\n</tool_call><|im_end|>\n<|im_start|>user\n<tool_response>\nMV2CTX memory: the user likes short answers\n</tool_response><|im_end|>\n<|im_start|>assistant\n<think>\n"
  },
  {
   "repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-7B/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "56a1447ad31926fd",
   "shape": "A_0130_system_prompt",
   "args": "dict",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "<s>You are a helpful assistant.\n\nMV2CTX memory: the user likes short answers<｜User｜>MV2USER hello, what do you remember about me?<｜Assistant｜><think>\n"
  },
  {
   "repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-7B/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "56a1447ad31926fd",
   "shape": "B_0131_hidden_tool",
   "args": "dict",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "<s>You are a helpful assistant.<｜User｜>MV2USER hello, what do you remember about me?<｜Assistant｜><｜end▁of▁sentence｜><｜tool▁outputs▁begin｜><｜tool▁output▁begin｜>MV2CTX memory: the user likes short answers<｜tool▁output▁end｜><｜tool▁outputs▁end｜>"
  },
  {
   "repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-7B/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "56a1447ad31926fd",
   "shape": "B_0131_hidden_tool",
   "args": "dict",
   "id": "uuid4",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "<s>You are a helpful assistant.<｜User｜>MV2USER hello, what do you remember about me?<｜Assistant｜><｜end▁of▁sentence｜><｜tool▁outputs▁begin｜><｜tool▁output▁begin｜>MV2CTX memory: the user likes short answers<｜tool▁output▁end｜><｜tool▁outputs▁end｜>"
  },
  {
   "repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-7B/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "56a1447ad31926fd",
   "shape": "B_0131_hidden_tool",
   "args": "dict",
   "id": "alnum9",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "<s>You are a helpful assistant.<｜User｜>MV2USER hello, what do you remember about me?<｜Assistant｜><｜end▁of▁sentence｜><｜tool▁outputs▁begin｜><｜tool▁output▁begin｜>MV2CTX memory: the user likes short answers<｜tool▁output▁end｜><｜tool▁outputs▁end｜>"
  },
  {
   "repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-7B/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "56a1447ad31926fd",
   "shape": "B_0131_hidden_tool",
   "args": "str",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "<s>You are a helpful assistant.<｜User｜>MV2USER hello, what do you remember about me?<｜Assistant｜><｜end▁of▁sentence｜><｜tool▁outputs▁begin｜><｜tool▁output▁begin｜>MV2CTX memory: the user likes short answers<｜tool▁output▁end｜><｜tool▁outputs▁end｜>"
  },
  {
   "repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-7B/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "56a1447ad31926fd",
   "shape": "B_0131_hidden_tool",
   "args": "str",
   "id": "uuid4",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "<s>You are a helpful assistant.<｜User｜>MV2USER hello, what do you remember about me?<｜Assistant｜><｜end▁of▁sentence｜><｜tool▁outputs▁begin｜><｜tool▁output▁begin｜>MV2CTX memory: the user likes short answers<｜tool▁output▁end｜><｜tool▁outputs▁end｜>"
  },
  {
   "repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-7B/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "56a1447ad31926fd",
   "shape": "B_0131_hidden_tool",
   "args": "str",
   "id": "alnum9",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "<s>You are a helpful assistant.<｜User｜>MV2USER hello, what do you remember about me?<｜Assistant｜><｜end▁of▁sentence｜><｜tool▁outputs▁begin｜><｜tool▁output▁begin｜>MV2CTX memory: the user likes short answers<｜tool▁output▁end｜><｜tool▁outputs▁end｜>"
  },
  {
   "repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-7B/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "56a1447ad31926fd",
   "shape": "C_0131_no_system",
   "args": "dict",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "<s><｜User｜>MV2USER hello, what do you remember about me?<｜Assistant｜><｜end▁of▁sentence｜><｜tool▁outputs▁begin｜><｜tool▁output▁begin｜>MV2CTX memory: the user likes short answers<｜tool▁output▁end｜><｜tool▁outputs▁end｜>"
  },
  {
   "repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-7B/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "56a1447ad31926fd",
   "shape": "D_0131_content_null",
   "args": "dict",
   "id": "call_hex24",
   "ok": false,
   "error_type": "TypeError",
   "error": "can only concatenate str (not \"dict\") to str"
  },
  {
   "repo": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-7B/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "56a1447ad31926fd",
   "shape": "E_0131_empty_user",
   "args": "dict",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": false,
   "ctx_in_prompt": true,
   "tail": "<s>You are a helpful assistant.<｜User｜><｜Assistant｜><｜end▁of▁sentence｜><｜tool▁outputs▁begin｜><｜tool▁output▁begin｜>MV2CTX memory: the user likes short answers<｜tool▁output▁end｜><｜tool▁outputs▁end｜>"
  },
  {
   "repo": "deepseek-ai/DeepSeek-V3.1",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/deepseek-ai/DeepSeek-V3.1/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "45690185067ea25c",
   "shape": "A_0130_system_prompt",
   "args": "dict",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "<s>You are a helpful assistant.\n\nMV2CTX memory: the user likes short answers<｜User｜>MV2USER hello, what do you remember about me?<｜Assistant｜></think>"
  },
  {
   "repo": "deepseek-ai/DeepSeek-V3.1",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/deepseek-ai/DeepSeek-V3.1/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "45690185067ea25c",
   "shape": "B_0131_hidden_tool",
   "args": "dict",
   "id": "call_hex24",
   "ok": false,
   "error_type": "TypeError",
   "error": "can only concatenate str (not \"dict\") to str"
  },
  {
   "repo": "deepseek-ai/DeepSeek-V3.1",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/deepseek-ai/DeepSeek-V3.1/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "45690185067ea25c",
   "shape": "B_0131_hidden_tool",
   "args": "dict",
   "id": "uuid4",
   "ok": false,
   "error_type": "TypeError",
   "error": "can only concatenate str (not \"dict\") to str"
  },
  {
   "repo": "deepseek-ai/DeepSeek-V3.1",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/deepseek-ai/DeepSeek-V3.1/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "45690185067ea25c",
   "shape": "B_0131_hidden_tool",
   "args": "dict",
   "id": "alnum9",
   "ok": false,
   "error_type": "TypeError",
   "error": "can only concatenate str (not \"dict\") to str"
  },
  {
   "repo": "deepseek-ai/DeepSeek-V3.1",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/deepseek-ai/DeepSeek-V3.1/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "45690185067ea25c",
   "shape": "B_0131_hidden_tool",
   "args": "str",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "<s>You are a helpful assistant.<｜User｜>MV2USER hello, what do you remember about me?<｜Assistant｜></think><｜tool▁calls▁begin｜><｜tool▁call▁begin｜>recall_context<｜tool▁sep｜>{\"query\": \"user\"}<｜tool▁call▁end｜><｜tool▁calls▁end｜><｜end▁of▁sentence｜><｜tool▁output▁begin｜>MV2CTX memory: the user likes short answers<｜tool▁output▁end｜>"
  },
  {
   "repo": "deepseek-ai/DeepSeek-V3.1",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/deepseek-ai/DeepSeek-V3.1/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "45690185067ea25c",
   "shape": "B_0131_hidden_tool",
   "args": "str",
   "id": "uuid4",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "<s>You are a helpful assistant.<｜User｜>MV2USER hello, what do you remember about me?<｜Assistant｜></think><｜tool▁calls▁begin｜><｜tool▁call▁begin｜>recall_context<｜tool▁sep｜>{\"query\": \"user\"}<｜tool▁call▁end｜><｜tool▁calls▁end｜><｜end▁of▁sentence｜><｜tool▁output▁begin｜>MV2CTX memory: the user likes short answers<｜tool▁output▁end｜>"
  },
  {
   "repo": "deepseek-ai/DeepSeek-V3.1",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/deepseek-ai/DeepSeek-V3.1/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "45690185067ea25c",
   "shape": "B_0131_hidden_tool",
   "args": "str",
   "id": "alnum9",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "<s>You are a helpful assistant.<｜User｜>MV2USER hello, what do you remember about me?<｜Assistant｜></think><｜tool▁calls▁begin｜><｜tool▁call▁begin｜>recall_context<｜tool▁sep｜>{\"query\": \"user\"}<｜tool▁call▁end｜><｜tool▁calls▁end｜><｜end▁of▁sentence｜><｜tool▁output▁begin｜>MV2CTX memory: the user likes short answers<｜tool▁output▁end｜>"
  },
  {
   "repo": "deepseek-ai/DeepSeek-V3.1",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/deepseek-ai/DeepSeek-V3.1/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "45690185067ea25c",
   "shape": "C_0131_no_system",
   "args": "dict",
   "id": "call_hex24",
   "ok": false,
   "error_type": "TypeError",
   "error": "can only concatenate str (not \"dict\") to str"
  },
  {
   "repo": "deepseek-ai/DeepSeek-V3.1",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/deepseek-ai/DeepSeek-V3.1/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "45690185067ea25c",
   "shape": "D_0131_content_null",
   "args": "dict",
   "id": "call_hex24",
   "ok": false,
   "error_type": "TypeError",
   "error": "can only concatenate str (not \"dict\") to str"
  },
  {
   "repo": "deepseek-ai/DeepSeek-V3.1",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/deepseek-ai/DeepSeek-V3.1/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "45690185067ea25c",
   "shape": "E_0131_empty_user",
   "args": "dict",
   "id": "call_hex24",
   "ok": false,
   "error_type": "TypeError",
   "error": "can only concatenate str (not \"dict\") to str"
  },
  {
   "repo": "deepseek-ai/DeepSeek-R1-0528",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/deepseek-ai/DeepSeek-R1-0528/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "53671bac29bcb3e1",
   "shape": "A_0130_system_prompt",
   "args": "dict",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "<s>You are a helpful assistant.\n\nMV2CTX memory: the user likes short answers<｜User｜>MV2USER hello, what do you remember about me?<｜Assistant｜>"
  },
  {
   "repo": "deepseek-ai/DeepSeek-R1-0528",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/deepseek-ai/DeepSeek-R1-0528/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "53671bac29bcb3e1",
   "shape": "B_0131_hidden_tool",
   "args": "dict",
   "id": "call_hex24",
   "ok": false,
   "error_type": "TypeError",
   "error": "can only concatenate str (not \"dict\") to str"
  },
  {
   "repo": "deepseek-ai/DeepSeek-R1-0528",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/deepseek-ai/DeepSeek-R1-0528/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "53671bac29bcb3e1",
   "shape": "B_0131_hidden_tool",
   "args": "dict",
   "id": "uuid4",
   "ok": false,
   "error_type": "TypeError",
   "error": "can only concatenate str (not \"dict\") to str"
  },
  {
   "repo": "deepseek-ai/DeepSeek-R1-0528",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/deepseek-ai/DeepSeek-R1-0528/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "53671bac29bcb3e1",
   "shape": "B_0131_hidden_tool",
   "args": "dict",
   "id": "alnum9",
   "ok": false,
   "error_type": "TypeError",
   "error": "can only concatenate str (not \"dict\") to str"
  },
  {
   "repo": "deepseek-ai/DeepSeek-R1-0528",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/deepseek-ai/DeepSeek-R1-0528/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "53671bac29bcb3e1",
   "shape": "B_0131_hidden_tool",
   "args": "str",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "<s>You are a helpful assistant.<｜User｜>MV2USER hello, what do you remember about me?<｜Assistant｜><｜tool▁calls▁begin｜><｜tool▁call▁begin｜>function<｜tool▁sep｜>recall_context\n```json\n{\"query\": \"user\"}\n```<｜tool▁call▁end｜><｜tool▁calls▁end｜><｜end▁of▁sentence｜><｜tool▁outputs▁begin｜><｜tool▁output▁begin｜>MV2CTX memory: the user likes short answers<｜tool▁output▁end｜><｜tool▁outputs▁end｜>"
  },
  {
   "repo": "deepseek-ai/DeepSeek-R1-0528",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/deepseek-ai/DeepSeek-R1-0528/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "53671bac29bcb3e1",
   "shape": "B_0131_hidden_tool",
   "args": "str",
   "id": "uuid4",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "<s>You are a helpful assistant.<｜User｜>MV2USER hello, what do you remember about me?<｜Assistant｜><｜tool▁calls▁begin｜><｜tool▁call▁begin｜>function<｜tool▁sep｜>recall_context\n```json\n{\"query\": \"user\"}\n```<｜tool▁call▁end｜><｜tool▁calls▁end｜><｜end▁of▁sentence｜><｜tool▁outputs▁begin｜><｜tool▁output▁begin｜>MV2CTX memory: the user likes short answers<｜tool▁output▁end｜><｜tool▁outputs▁end｜>"
  },
  {
   "repo": "deepseek-ai/DeepSeek-R1-0528",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/deepseek-ai/DeepSeek-R1-0528/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "53671bac29bcb3e1",
   "shape": "B_0131_hidden_tool",
   "args": "str",
   "id": "alnum9",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "<s>You are a helpful assistant.<｜User｜>MV2USER hello, what do you remember about me?<｜Assistant｜><｜tool▁calls▁begin｜><｜tool▁call▁begin｜>function<｜tool▁sep｜>recall_context\n```json\n{\"query\": \"user\"}\n```<｜tool▁call▁end｜><｜tool▁calls▁end｜><｜end▁of▁sentence｜><｜tool▁outputs▁begin｜><｜tool▁output▁begin｜>MV2CTX memory: the user likes short answers<｜tool▁output▁end｜><｜tool▁outputs▁end｜>"
  },
  {
   "repo": "deepseek-ai/DeepSeek-R1-0528",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/deepseek-ai/DeepSeek-R1-0528/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "53671bac29bcb3e1",
   "shape": "C_0131_no_system",
   "args": "dict",
   "id": "call_hex24",
   "ok": false,
   "error_type": "TypeError",
   "error": "can only concatenate str (not \"dict\") to str"
  },
  {
   "repo": "deepseek-ai/DeepSeek-R1-0528",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/deepseek-ai/DeepSeek-R1-0528/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "53671bac29bcb3e1",
   "shape": "D_0131_content_null",
   "args": "dict",
   "id": "call_hex24",
   "ok": false,
   "error_type": "TypeError",
   "error": "argument of type 'NoneType' is not iterable"
  },
  {
   "repo": "deepseek-ai/DeepSeek-R1-0528",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/deepseek-ai/DeepSeek-R1-0528/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "53671bac29bcb3e1",
   "shape": "E_0131_empty_user",
   "args": "dict",
   "id": "call_hex24",
   "ok": false,
   "error_type": "TypeError",
   "error": "can only concatenate str (not \"dict\") to str"
  },
  {
   "repo": "unsloth/gemma-3-4b-it",
   "fetched": true,
   "file": "chat_template.jinja",
   "url": "https://huggingface.co/unsloth/gemma-3-4b-it/resolve/main/chat_template.jinja",
   "template": "default",
   "sha16": "7de1c58e208eda46",
   "shape": "A_0130_system_prompt",
   "args": "dict",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "<s><start_of_turn>user\nYou are a helpful assistant.\n\nMV2CTX memory: the user likes short answers\n\nMV2USER hello, what do you remember about me?<end_of_turn>\n<start_of_turn>model\n"
  },
  {
   "repo": "unsloth/gemma-3-4b-it",
   "fetched": true,
   "file": "chat_template.jinja",
   "url": "https://huggingface.co/unsloth/gemma-3-4b-it/resolve/main/chat_template.jinja",
   "template": "default",
   "sha16": "7de1c58e208eda46",
   "shape": "B_0131_hidden_tool",
   "args": "dict",
   "id": "call_hex24",
   "ok": false,
   "error_type": "TemplateError",
   "error": "Conversation roles must alternate user/assistant/user/assistant/..."
  },
  {
   "repo": "unsloth/gemma-3-4b-it",
   "fetched": true,
   "file": "chat_template.jinja",
   "url": "https://huggingface.co/unsloth/gemma-3-4b-it/resolve/main/chat_template.jinja",
   "template": "default",
   "sha16": "7de1c58e208eda46",
   "shape": "B_0131_hidden_tool",
   "args": "dict",
   "id": "uuid4",
   "ok": false,
   "error_type": "TemplateError",
   "error": "Conversation roles must alternate user/assistant/user/assistant/..."
  },
  {
   "repo": "unsloth/gemma-3-4b-it",
   "fetched": true,
   "file": "chat_template.jinja",
   "url": "https://huggingface.co/unsloth/gemma-3-4b-it/resolve/main/chat_template.jinja",
   "template": "default",
   "sha16": "7de1c58e208eda46",
   "shape": "B_0131_hidden_tool",
   "args": "dict",
   "id": "alnum9",
   "ok": false,
   "error_type": "TemplateError",
   "error": "Conversation roles must alternate user/assistant/user/assistant/..."
  },
  {
   "repo": "unsloth/gemma-3-4b-it",
   "fetched": true,
   "file": "chat_template.jinja",
   "url": "https://huggingface.co/unsloth/gemma-3-4b-it/resolve/main/chat_template.jinja",
   "template": "default",
   "sha16": "7de1c58e208eda46",
   "shape": "B_0131_hidden_tool",
   "args": "str",
   "id": "call_hex24",
   "ok": false,
   "error_type": "TemplateError",
   "error": "Conversation roles must alternate user/assistant/user/assistant/..."
  },
  {
   "repo": "unsloth/gemma-3-4b-it",
   "fetched": true,
   "file": "chat_template.jinja",
   "url": "https://huggingface.co/unsloth/gemma-3-4b-it/resolve/main/chat_template.jinja",
   "template": "default",
   "sha16": "7de1c58e208eda46",
   "shape": "B_0131_hidden_tool",
   "args": "str",
   "id": "uuid4",
   "ok": false,
   "error_type": "TemplateError",
   "error": "Conversation roles must alternate user/assistant/user/assistant/..."
  },
  {
   "repo": "unsloth/gemma-3-4b-it",
   "fetched": true,
   "file": "chat_template.jinja",
   "url": "https://huggingface.co/unsloth/gemma-3-4b-it/resolve/main/chat_template.jinja",
   "template": "default",
   "sha16": "7de1c58e208eda46",
   "shape": "B_0131_hidden_tool",
   "args": "str",
   "id": "alnum9",
   "ok": false,
   "error_type": "TemplateError",
   "error": "Conversation roles must alternate user/assistant/user/assistant/..."
  },
  {
   "repo": "unsloth/gemma-3-4b-it",
   "fetched": true,
   "file": "chat_template.jinja",
   "url": "https://huggingface.co/unsloth/gemma-3-4b-it/resolve/main/chat_template.jinja",
   "template": "default",
   "sha16": "7de1c58e208eda46",
   "shape": "C_0131_no_system",
   "args": "dict",
   "id": "call_hex24",
   "ok": false,
   "error_type": "TemplateError",
   "error": "Conversation roles must alternate user/assistant/user/assistant/..."
  },
  {
   "repo": "unsloth/gemma-3-4b-it",
   "fetched": true,
   "file": "chat_template.jinja",
   "url": "https://huggingface.co/unsloth/gemma-3-4b-it/resolve/main/chat_template.jinja",
   "template": "default",
   "sha16": "7de1c58e208eda46",
   "shape": "D_0131_content_null",
   "args": "dict",
   "id": "call_hex24",
   "ok": false,
   "error_type": "TemplateError",
   "error": "Invalid content type"
  },
  {
   "repo": "unsloth/gemma-3-4b-it",
   "fetched": true,
   "file": "chat_template.jinja",
   "url": "https://huggingface.co/unsloth/gemma-3-4b-it/resolve/main/chat_template.jinja",
   "template": "default",
   "sha16": "7de1c58e208eda46",
   "shape": "E_0131_empty_user",
   "args": "dict",
   "id": "call_hex24",
   "ok": false,
   "error_type": "TemplateError",
   "error": "Conversation roles must alternate user/assistant/user/assistant/..."
  },
  {
   "repo": "unsloth/gemma-3n-E4B-it",
   "fetched": true,
   "file": "chat_template.jinja",
   "url": "https://huggingface.co/unsloth/gemma-3n-E4B-it/resolve/main/chat_template.jinja",
   "template": "default",
   "sha16": "ac03dcb3b09726f1",
   "shape": "A_0130_system_prompt",
   "args": "dict",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "<s><start_of_turn>user\nYou are a helpful assistant.\n\nMV2CTX memory: the user likes short answers\n\nMV2USER hello, what do you remember about me?<end_of_turn>\n<start_of_turn>model\n"
  },
  {
   "repo": "unsloth/gemma-3n-E4B-it",
   "fetched": true,
   "file": "chat_template.jinja",
   "url": "https://huggingface.co/unsloth/gemma-3n-E4B-it/resolve/main/chat_template.jinja",
   "template": "default",
   "sha16": "ac03dcb3b09726f1",
   "shape": "B_0131_hidden_tool",
   "args": "dict",
   "id": "call_hex24",
   "ok": false,
   "error_type": "TemplateError",
   "error": "Conversation roles must alternate user/assistant/user/assistant/..."
  },
  {
   "repo": "unsloth/gemma-3n-E4B-it",
   "fetched": true,
   "file": "chat_template.jinja",
   "url": "https://huggingface.co/unsloth/gemma-3n-E4B-it/resolve/main/chat_template.jinja",
   "template": "default",
   "sha16": "ac03dcb3b09726f1",
   "shape": "B_0131_hidden_tool",
   "args": "dict",
   "id": "uuid4",
   "ok": false,
   "error_type": "TemplateError",
   "error": "Conversation roles must alternate user/assistant/user/assistant/..."
  },
  {
   "repo": "unsloth/gemma-3n-E4B-it",
   "fetched": true,
   "file": "chat_template.jinja",
   "url": "https://huggingface.co/unsloth/gemma-3n-E4B-it/resolve/main/chat_template.jinja",
   "template": "default",
   "sha16": "ac03dcb3b09726f1",
   "shape": "B_0131_hidden_tool",
   "args": "dict",
   "id": "alnum9",
   "ok": false,
   "error_type": "TemplateError",
   "error": "Conversation roles must alternate user/assistant/user/assistant/..."
  },
  {
   "repo": "unsloth/gemma-3n-E4B-it",
   "fetched": true,
   "file": "chat_template.jinja",
   "url": "https://huggingface.co/unsloth/gemma-3n-E4B-it/resolve/main/chat_template.jinja",
   "template": "default",
   "sha16": "ac03dcb3b09726f1",
   "shape": "B_0131_hidden_tool",
   "args": "str",
   "id": "call_hex24",
   "ok": false,
   "error_type": "TemplateError",
   "error": "Conversation roles must alternate user/assistant/user/assistant/..."
  },
  {
   "repo": "unsloth/gemma-3n-E4B-it",
   "fetched": true,
   "file": "chat_template.jinja",
   "url": "https://huggingface.co/unsloth/gemma-3n-E4B-it/resolve/main/chat_template.jinja",
   "template": "default",
   "sha16": "ac03dcb3b09726f1",
   "shape": "B_0131_hidden_tool",
   "args": "str",
   "id": "uuid4",
   "ok": false,
   "error_type": "TemplateError",
   "error": "Conversation roles must alternate user/assistant/user/assistant/..."
  },
  {
   "repo": "unsloth/gemma-3n-E4B-it",
   "fetched": true,
   "file": "chat_template.jinja",
   "url": "https://huggingface.co/unsloth/gemma-3n-E4B-it/resolve/main/chat_template.jinja",
   "template": "default",
   "sha16": "ac03dcb3b09726f1",
   "shape": "B_0131_hidden_tool",
   "args": "str",
   "id": "alnum9",
   "ok": false,
   "error_type": "TemplateError",
   "error": "Conversation roles must alternate user/assistant/user/assistant/..."
  },
  {
   "repo": "unsloth/gemma-3n-E4B-it",
   "fetched": true,
   "file": "chat_template.jinja",
   "url": "https://huggingface.co/unsloth/gemma-3n-E4B-it/resolve/main/chat_template.jinja",
   "template": "default",
   "sha16": "ac03dcb3b09726f1",
   "shape": "C_0131_no_system",
   "args": "dict",
   "id": "call_hex24",
   "ok": false,
   "error_type": "TemplateError",
   "error": "Conversation roles must alternate user/assistant/user/assistant/..."
  },
  {
   "repo": "unsloth/gemma-3n-E4B-it",
   "fetched": true,
   "file": "chat_template.jinja",
   "url": "https://huggingface.co/unsloth/gemma-3n-E4B-it/resolve/main/chat_template.jinja",
   "template": "default",
   "sha16": "ac03dcb3b09726f1",
   "shape": "D_0131_content_null",
   "args": "dict",
   "id": "call_hex24",
   "ok": false,
   "error_type": "TemplateError",
   "error": "Invalid content type"
  },
  {
   "repo": "unsloth/gemma-3n-E4B-it",
   "fetched": true,
   "file": "chat_template.jinja",
   "url": "https://huggingface.co/unsloth/gemma-3n-E4B-it/resolve/main/chat_template.jinja",
   "template": "default",
   "sha16": "ac03dcb3b09726f1",
   "shape": "E_0131_empty_user",
   "args": "dict",
   "id": "call_hex24",
   "ok": false,
   "error_type": "TemplateError",
   "error": "Conversation roles must alternate user/assistant/user/assistant/..."
  },
  {
   "repo": "unsloth/Llama-3.1-8B-Instruct",
   "fetched": true,
   "file": "chat_template.jinja",
   "url": "https://huggingface.co/unsloth/Llama-3.1-8B-Instruct/resolve/main/chat_template.jinja",
   "template": "default",
   "sha16": "e10ca381b1ccc5cf",
   "shape": "A_0130_system_prompt",
   "args": "dict",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "unctions, please respond with a JSON for a function call with its proper arguments that best answers the given prompt.\n\nRespond in the format {\"name\": function name, \"parameters\": dictionary of argument name and its value}.Do not use variables.\n\n{\n    \"type\": \"function\",\n    \"function\": {\n        \"name\": \"recall_context\",\n        \"description\": \"Recall saved memories.\",\n        \"parameters\": {\n            \"type\": \"object\",\n            \"properties\": {\n                \"query\": {\n                    \"type\": \"string\"\n                }\n            },\n            \"required\": []\n        }\n    }\n}\n\nMV2USER hello, what do you remember about me?<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n"
  },
  {
   "repo": "unsloth/Llama-3.1-8B-Instruct",
   "fetched": true,
   "file": "chat_template.jinja",
   "url": "https://huggingface.co/unsloth/Llama-3.1-8B-Instruct/resolve/main/chat_template.jinja",
   "template": "default",
   "sha16": "e10ca381b1ccc5cf",
   "shape": "B_0131_hidden_tool",
   "args": "dict",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "value}.Do not use variables.\n\n{\n    \"type\": \"function\",\n    \"function\": {\n        \"name\": \"recall_context\",\n        \"description\": \"Recall saved memories.\",\n        \"parameters\": {\n            \"type\": \"object\",\n            \"properties\": {\n                \"query\": {\n                    \"type\": \"string\"\n                }\n            },\n            \"required\": []\n        }\n    }\n}\n\nMV2USER hello, what do you remember about me?<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n{\"name\": \"recall_context\", \"parameters\": {\"query\": \"user\"}}<|eot_id|><|start_header_id|>ipython<|end_header_id|>\n\n\"MV2CTX memory: the user likes short answers\"<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n"
  },
  {
   "repo": "unsloth/Llama-3.1-8B-Instruct",
   "fetched": true,
   "file": "chat_template.jinja",
   "url": "https://huggingface.co/unsloth/Llama-3.1-8B-Instruct/resolve/main/chat_template.jinja",
   "template": "default",
   "sha16": "e10ca381b1ccc5cf",
   "shape": "B_0131_hidden_tool",
   "args": "dict",
   "id": "uuid4",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "value}.Do not use variables.\n\n{\n    \"type\": \"function\",\n    \"function\": {\n        \"name\": \"recall_context\",\n        \"description\": \"Recall saved memories.\",\n        \"parameters\": {\n            \"type\": \"object\",\n            \"properties\": {\n                \"query\": {\n                    \"type\": \"string\"\n                }\n            },\n            \"required\": []\n        }\n    }\n}\n\nMV2USER hello, what do you remember about me?<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n{\"name\": \"recall_context\", \"parameters\": {\"query\": \"user\"}}<|eot_id|><|start_header_id|>ipython<|end_header_id|>\n\n\"MV2CTX memory: the user likes short answers\"<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n"
  },
  {
   "repo": "unsloth/Llama-3.1-8B-Instruct",
   "fetched": true,
   "file": "chat_template.jinja",
   "url": "https://huggingface.co/unsloth/Llama-3.1-8B-Instruct/resolve/main/chat_template.jinja",
   "template": "default",
   "sha16": "e10ca381b1ccc5cf",
   "shape": "B_0131_hidden_tool",
   "args": "dict",
   "id": "alnum9",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "value}.Do not use variables.\n\n{\n    \"type\": \"function\",\n    \"function\": {\n        \"name\": \"recall_context\",\n        \"description\": \"Recall saved memories.\",\n        \"parameters\": {\n            \"type\": \"object\",\n            \"properties\": {\n                \"query\": {\n                    \"type\": \"string\"\n                }\n            },\n            \"required\": []\n        }\n    }\n}\n\nMV2USER hello, what do you remember about me?<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n{\"name\": \"recall_context\", \"parameters\": {\"query\": \"user\"}}<|eot_id|><|start_header_id|>ipython<|end_header_id|>\n\n\"MV2CTX memory: the user likes short answers\"<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n"
  },
  {
   "repo": "unsloth/Llama-3.1-8B-Instruct",
   "fetched": true,
   "file": "chat_template.jinja",
   "url": "https://huggingface.co/unsloth/Llama-3.1-8B-Instruct/resolve/main/chat_template.jinja",
   "template": "default",
   "sha16": "e10ca381b1ccc5cf",
   "shape": "B_0131_hidden_tool",
   "args": "str",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": ".Do not use variables.\n\n{\n    \"type\": \"function\",\n    \"function\": {\n        \"name\": \"recall_context\",\n        \"description\": \"Recall saved memories.\",\n        \"parameters\": {\n            \"type\": \"object\",\n            \"properties\": {\n                \"query\": {\n                    \"type\": \"string\"\n                }\n            },\n            \"required\": []\n        }\n    }\n}\n\nMV2USER hello, what do you remember about me?<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n{\"name\": \"recall_context\", \"parameters\": \"{\\\"query\\\": \\\"user\\\"}\"}<|eot_id|><|start_header_id|>ipython<|end_header_id|>\n\n\"MV2CTX memory: the user likes short answers\"<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n"
  },
  {
   "repo": "unsloth/Llama-3.1-8B-Instruct",
   "fetched": true,
   "file": "chat_template.jinja",
   "url": "https://huggingface.co/unsloth/Llama-3.1-8B-Instruct/resolve/main/chat_template.jinja",
   "template": "default",
   "sha16": "e10ca381b1ccc5cf",
   "shape": "B_0131_hidden_tool",
   "args": "str",
   "id": "uuid4",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": ".Do not use variables.\n\n{\n    \"type\": \"function\",\n    \"function\": {\n        \"name\": \"recall_context\",\n        \"description\": \"Recall saved memories.\",\n        \"parameters\": {\n            \"type\": \"object\",\n            \"properties\": {\n                \"query\": {\n                    \"type\": \"string\"\n                }\n            },\n            \"required\": []\n        }\n    }\n}\n\nMV2USER hello, what do you remember about me?<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n{\"name\": \"recall_context\", \"parameters\": \"{\\\"query\\\": \\\"user\\\"}\"}<|eot_id|><|start_header_id|>ipython<|end_header_id|>\n\n\"MV2CTX memory: the user likes short answers\"<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n"
  },
  {
   "repo": "unsloth/Llama-3.1-8B-Instruct",
   "fetched": true,
   "file": "chat_template.jinja",
   "url": "https://huggingface.co/unsloth/Llama-3.1-8B-Instruct/resolve/main/chat_template.jinja",
   "template": "default",
   "sha16": "e10ca381b1ccc5cf",
   "shape": "B_0131_hidden_tool",
   "args": "str",
   "id": "alnum9",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": ".Do not use variables.\n\n{\n    \"type\": \"function\",\n    \"function\": {\n        \"name\": \"recall_context\",\n        \"description\": \"Recall saved memories.\",\n        \"parameters\": {\n            \"type\": \"object\",\n            \"properties\": {\n                \"query\": {\n                    \"type\": \"string\"\n                }\n            },\n            \"required\": []\n        }\n    }\n}\n\nMV2USER hello, what do you remember about me?<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n{\"name\": \"recall_context\", \"parameters\": \"{\\\"query\\\": \\\"user\\\"}\"}<|eot_id|><|start_header_id|>ipython<|end_header_id|>\n\n\"MV2CTX memory: the user likes short answers\"<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n"
  },
  {
   "repo": "unsloth/Llama-3.1-8B-Instruct",
   "fetched": true,
   "file": "chat_template.jinja",
   "url": "https://huggingface.co/unsloth/Llama-3.1-8B-Instruct/resolve/main/chat_template.jinja",
   "template": "default",
   "sha16": "e10ca381b1ccc5cf",
   "shape": "C_0131_no_system",
   "args": "dict",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "value}.Do not use variables.\n\n{\n    \"type\": \"function\",\n    \"function\": {\n        \"name\": \"recall_context\",\n        \"description\": \"Recall saved memories.\",\n        \"parameters\": {\n            \"type\": \"object\",\n            \"properties\": {\n                \"query\": {\n                    \"type\": \"string\"\n                }\n            },\n            \"required\": []\n        }\n    }\n}\n\nMV2USER hello, what do you remember about me?<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n{\"name\": \"recall_context\", \"parameters\": {\"query\": \"user\"}}<|eot_id|><|start_header_id|>ipython<|end_header_id|>\n\n\"MV2CTX memory: the user likes short answers\"<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n"
  },
  {
   "repo": "unsloth/Llama-3.1-8B-Instruct",
   "fetched": true,
   "file": "chat_template.jinja",
   "url": "https://huggingface.co/unsloth/Llama-3.1-8B-Instruct/resolve/main/chat_template.jinja",
   "template": "default",
   "sha16": "e10ca381b1ccc5cf",
   "shape": "D_0131_content_null",
   "args": "dict",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "value}.Do not use variables.\n\n{\n    \"type\": \"function\",\n    \"function\": {\n        \"name\": \"recall_context\",\n        \"description\": \"Recall saved memories.\",\n        \"parameters\": {\n            \"type\": \"object\",\n            \"properties\": {\n                \"query\": {\n                    \"type\": \"string\"\n                }\n            },\n            \"required\": []\n        }\n    }\n}\n\nMV2USER hello, what do you remember about me?<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n{\"name\": \"recall_context\", \"parameters\": {\"query\": \"user\"}}<|eot_id|><|start_header_id|>ipython<|end_header_id|>\n\n\"MV2CTX memory: the user likes short answers\"<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n"
  },
  {
   "repo": "unsloth/Llama-3.1-8B-Instruct",
   "fetched": true,
   "file": "chat_template.jinja",
   "url": "https://huggingface.co/unsloth/Llama-3.1-8B-Instruct/resolve/main/chat_template.jinja",
   "template": "default",
   "sha16": "e10ca381b1ccc5cf",
   "shape": "E_0131_empty_user",
   "args": "dict",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": false,
   "ctx_in_prompt": true,
   "tail": "meters\": dictionary of argument name and its value}.Do not use variables.\n\n{\n    \"type\": \"function\",\n    \"function\": {\n        \"name\": \"recall_context\",\n        \"description\": \"Recall saved memories.\",\n        \"parameters\": {\n            \"type\": \"object\",\n            \"properties\": {\n                \"query\": {\n                    \"type\": \"string\"\n                }\n            },\n            \"required\": []\n        }\n    }\n}\n\n<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n{\"name\": \"recall_context\", \"parameters\": {\"query\": \"user\"}}<|eot_id|><|start_header_id|>ipython<|end_header_id|>\n\n\"MV2CTX memory: the user likes short answers\"<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n"
  },
  {
   "repo": "unsloth/Llama-3.2-3B-Instruct",
   "fetched": true,
   "file": "chat_template.jinja",
   "url": "https://huggingface.co/unsloth/Llama-3.2-3B-Instruct/resolve/main/chat_template.jinja",
   "template": "default",
   "sha16": "5816fce10444e03c",
   "shape": "A_0130_system_prompt",
   "args": "dict",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "unctions, please respond with a JSON for a function call with its proper arguments that best answers the given prompt.\n\nRespond in the format {\"name\": function name, \"parameters\": dictionary of argument name and its value}.Do not use variables.\n\n{\n    \"type\": \"function\",\n    \"function\": {\n        \"name\": \"recall_context\",\n        \"description\": \"Recall saved memories.\",\n        \"parameters\": {\n            \"type\": \"object\",\n            \"properties\": {\n                \"query\": {\n                    \"type\": \"string\"\n                }\n            },\n            \"required\": []\n        }\n    }\n}\n\nMV2USER hello, what do you remember about me?<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n"
  },
  {
   "repo": "unsloth/Llama-3.2-3B-Instruct",
   "fetched": true,
   "file": "chat_template.jinja",
   "url": "https://huggingface.co/unsloth/Llama-3.2-3B-Instruct/resolve/main/chat_template.jinja",
   "template": "default",
   "sha16": "5816fce10444e03c",
   "shape": "B_0131_hidden_tool",
   "args": "dict",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "value}.Do not use variables.\n\n{\n    \"type\": \"function\",\n    \"function\": {\n        \"name\": \"recall_context\",\n        \"description\": \"Recall saved memories.\",\n        \"parameters\": {\n            \"type\": \"object\",\n            \"properties\": {\n                \"query\": {\n                    \"type\": \"string\"\n                }\n            },\n            \"required\": []\n        }\n    }\n}\n\nMV2USER hello, what do you remember about me?<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n{\"name\": \"recall_context\", \"parameters\": {\"query\": \"user\"}}<|eot_id|><|start_header_id|>ipython<|end_header_id|>\n\n\"MV2CTX memory: the user likes short answers\"<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n"
  },
  {
   "repo": "unsloth/Llama-3.2-3B-Instruct",
   "fetched": true,
   "file": "chat_template.jinja",
   "url": "https://huggingface.co/unsloth/Llama-3.2-3B-Instruct/resolve/main/chat_template.jinja",
   "template": "default",
   "sha16": "5816fce10444e03c",
   "shape": "B_0131_hidden_tool",
   "args": "dict",
   "id": "uuid4",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "value}.Do not use variables.\n\n{\n    \"type\": \"function\",\n    \"function\": {\n        \"name\": \"recall_context\",\n        \"description\": \"Recall saved memories.\",\n        \"parameters\": {\n            \"type\": \"object\",\n            \"properties\": {\n                \"query\": {\n                    \"type\": \"string\"\n                }\n            },\n            \"required\": []\n        }\n    }\n}\n\nMV2USER hello, what do you remember about me?<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n{\"name\": \"recall_context\", \"parameters\": {\"query\": \"user\"}}<|eot_id|><|start_header_id|>ipython<|end_header_id|>\n\n\"MV2CTX memory: the user likes short answers\"<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n"
  },
  {
   "repo": "unsloth/Llama-3.2-3B-Instruct",
   "fetched": true,
   "file": "chat_template.jinja",
   "url": "https://huggingface.co/unsloth/Llama-3.2-3B-Instruct/resolve/main/chat_template.jinja",
   "template": "default",
   "sha16": "5816fce10444e03c",
   "shape": "B_0131_hidden_tool",
   "args": "dict",
   "id": "alnum9",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "value}.Do not use variables.\n\n{\n    \"type\": \"function\",\n    \"function\": {\n        \"name\": \"recall_context\",\n        \"description\": \"Recall saved memories.\",\n        \"parameters\": {\n            \"type\": \"object\",\n            \"properties\": {\n                \"query\": {\n                    \"type\": \"string\"\n                }\n            },\n            \"required\": []\n        }\n    }\n}\n\nMV2USER hello, what do you remember about me?<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n{\"name\": \"recall_context\", \"parameters\": {\"query\": \"user\"}}<|eot_id|><|start_header_id|>ipython<|end_header_id|>\n\n\"MV2CTX memory: the user likes short answers\"<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n"
  },
  {
   "repo": "unsloth/Llama-3.2-3B-Instruct",
   "fetched": true,
   "file": "chat_template.jinja",
   "url": "https://huggingface.co/unsloth/Llama-3.2-3B-Instruct/resolve/main/chat_template.jinja",
   "template": "default",
   "sha16": "5816fce10444e03c",
   "shape": "B_0131_hidden_tool",
   "args": "str",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": ".Do not use variables.\n\n{\n    \"type\": \"function\",\n    \"function\": {\n        \"name\": \"recall_context\",\n        \"description\": \"Recall saved memories.\",\n        \"parameters\": {\n            \"type\": \"object\",\n            \"properties\": {\n                \"query\": {\n                    \"type\": \"string\"\n                }\n            },\n            \"required\": []\n        }\n    }\n}\n\nMV2USER hello, what do you remember about me?<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n{\"name\": \"recall_context\", \"parameters\": \"{\\\"query\\\": \\\"user\\\"}\"}<|eot_id|><|start_header_id|>ipython<|end_header_id|>\n\n\"MV2CTX memory: the user likes short answers\"<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n"
  },
  {
   "repo": "unsloth/Llama-3.2-3B-Instruct",
   "fetched": true,
   "file": "chat_template.jinja",
   "url": "https://huggingface.co/unsloth/Llama-3.2-3B-Instruct/resolve/main/chat_template.jinja",
   "template": "default",
   "sha16": "5816fce10444e03c",
   "shape": "B_0131_hidden_tool",
   "args": "str",
   "id": "uuid4",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": ".Do not use variables.\n\n{\n    \"type\": \"function\",\n    \"function\": {\n        \"name\": \"recall_context\",\n        \"description\": \"Recall saved memories.\",\n        \"parameters\": {\n            \"type\": \"object\",\n            \"properties\": {\n                \"query\": {\n                    \"type\": \"string\"\n                }\n            },\n            \"required\": []\n        }\n    }\n}\n\nMV2USER hello, what do you remember about me?<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n{\"name\": \"recall_context\", \"parameters\": \"{\\\"query\\\": \\\"user\\\"}\"}<|eot_id|><|start_header_id|>ipython<|end_header_id|>\n\n\"MV2CTX memory: the user likes short answers\"<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n"
  },
  {
   "repo": "unsloth/Llama-3.2-3B-Instruct",
   "fetched": true,
   "file": "chat_template.jinja",
   "url": "https://huggingface.co/unsloth/Llama-3.2-3B-Instruct/resolve/main/chat_template.jinja",
   "template": "default",
   "sha16": "5816fce10444e03c",
   "shape": "B_0131_hidden_tool",
   "args": "str",
   "id": "alnum9",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": ".Do not use variables.\n\n{\n    \"type\": \"function\",\n    \"function\": {\n        \"name\": \"recall_context\",\n        \"description\": \"Recall saved memories.\",\n        \"parameters\": {\n            \"type\": \"object\",\n            \"properties\": {\n                \"query\": {\n                    \"type\": \"string\"\n                }\n            },\n            \"required\": []\n        }\n    }\n}\n\nMV2USER hello, what do you remember about me?<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n{\"name\": \"recall_context\", \"parameters\": \"{\\\"query\\\": \\\"user\\\"}\"}<|eot_id|><|start_header_id|>ipython<|end_header_id|>\n\n\"MV2CTX memory: the user likes short answers\"<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n"
  },
  {
   "repo": "unsloth/Llama-3.2-3B-Instruct",
   "fetched": true,
   "file": "chat_template.jinja",
   "url": "https://huggingface.co/unsloth/Llama-3.2-3B-Instruct/resolve/main/chat_template.jinja",
   "template": "default",
   "sha16": "5816fce10444e03c",
   "shape": "C_0131_no_system",
   "args": "dict",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "value}.Do not use variables.\n\n{\n    \"type\": \"function\",\n    \"function\": {\n        \"name\": \"recall_context\",\n        \"description\": \"Recall saved memories.\",\n        \"parameters\": {\n            \"type\": \"object\",\n            \"properties\": {\n                \"query\": {\n                    \"type\": \"string\"\n                }\n            },\n            \"required\": []\n        }\n    }\n}\n\nMV2USER hello, what do you remember about me?<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n{\"name\": \"recall_context\", \"parameters\": {\"query\": \"user\"}}<|eot_id|><|start_header_id|>ipython<|end_header_id|>\n\n\"MV2CTX memory: the user likes short answers\"<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n"
  },
  {
   "repo": "unsloth/Llama-3.2-3B-Instruct",
   "fetched": true,
   "file": "chat_template.jinja",
   "url": "https://huggingface.co/unsloth/Llama-3.2-3B-Instruct/resolve/main/chat_template.jinja",
   "template": "default",
   "sha16": "5816fce10444e03c",
   "shape": "D_0131_content_null",
   "args": "dict",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "value}.Do not use variables.\n\n{\n    \"type\": \"function\",\n    \"function\": {\n        \"name\": \"recall_context\",\n        \"description\": \"Recall saved memories.\",\n        \"parameters\": {\n            \"type\": \"object\",\n            \"properties\": {\n                \"query\": {\n                    \"type\": \"string\"\n                }\n            },\n            \"required\": []\n        }\n    }\n}\n\nMV2USER hello, what do you remember about me?<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n{\"name\": \"recall_context\", \"parameters\": {\"query\": \"user\"}}<|eot_id|><|start_header_id|>ipython<|end_header_id|>\n\n\"MV2CTX memory: the user likes short answers\"<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n"
  },
  {
   "repo": "unsloth/Llama-3.2-3B-Instruct",
   "fetched": true,
   "file": "chat_template.jinja",
   "url": "https://huggingface.co/unsloth/Llama-3.2-3B-Instruct/resolve/main/chat_template.jinja",
   "template": "default",
   "sha16": "5816fce10444e03c",
   "shape": "E_0131_empty_user",
   "args": "dict",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": false,
   "ctx_in_prompt": true,
   "tail": "meters\": dictionary of argument name and its value}.Do not use variables.\n\n{\n    \"type\": \"function\",\n    \"function\": {\n        \"name\": \"recall_context\",\n        \"description\": \"Recall saved memories.\",\n        \"parameters\": {\n            \"type\": \"object\",\n            \"properties\": {\n                \"query\": {\n                    \"type\": \"string\"\n                }\n            },\n            \"required\": []\n        }\n    }\n}\n\n<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n{\"name\": \"recall_context\", \"parameters\": {\"query\": \"user\"}}<|eot_id|><|start_header_id|>ipython<|end_header_id|>\n\n\"MV2CTX memory: the user likes short answers\"<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n"
  },
  {
   "repo": "unsloth/mistral-7b-instruct-v0.3",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/unsloth/mistral-7b-instruct-v0.3/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "e16746b40344d6c5",
   "shape": "A_0130_system_prompt",
   "args": "dict",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "<s>[AVAILABLE_TOOLS] [{\"type\": \"function\", \"function\": {\"name\": \"recall_context\", \"description\": \"Recall saved memories.\", \"parameters\": {\"type\": \"object\", \"properties\": {\"query\": {\"type\": \"string\"}}, \"required\": []}}}][/AVAILABLE_TOOLS][INST] You are a helpful assistant.\n\nMV2CTX memory: the user likes short answers\n\nMV2USER hello, what do you remember about me?[/INST]"
  },
  {
   "repo": "unsloth/mistral-7b-instruct-v0.3",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/unsloth/mistral-7b-instruct-v0.3/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "e16746b40344d6c5",
   "shape": "B_0131_hidden_tool",
   "args": "dict",
   "id": "call_hex24",
   "ok": false,
   "error_type": "TemplateError",
   "error": "Tool call IDs should be alphanumeric strings with length 9!"
  },
  {
   "repo": "unsloth/mistral-7b-instruct-v0.3",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/unsloth/mistral-7b-instruct-v0.3/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "e16746b40344d6c5",
   "shape": "B_0131_hidden_tool",
   "args": "dict",
   "id": "uuid4",
   "ok": false,
   "error_type": "TemplateError",
   "error": "Tool call IDs should be alphanumeric strings with length 9!"
  },
  {
   "repo": "unsloth/mistral-7b-instruct-v0.3",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/unsloth/mistral-7b-instruct-v0.3/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "e16746b40344d6c5",
   "shape": "B_0131_hidden_tool",
   "args": "dict",
   "id": "alnum9",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "<s>[AVAILABLE_TOOLS] [{\"type\": \"function\", \"function\": {\"name\": \"recall_context\", \"description\": \"Recall saved memories.\", \"parameters\": {\"type\": \"object\", \"properties\": {\"query\": {\"type\": \"string\"}}, \"required\": []}}}][/AVAILABLE_TOOLS][INST] MV2USER hello, what do you remember about me?[/INST][TOOL_CALLS] [{\"name\": \"recall_context\", \"arguments\": {\"query\": \"user\"}, \"id\": \"a1B2c3D4e\"}]</s>[TOOL_RESULTS] {\"content\": MV2CTX memory: the user likes short answers, \"call_id\": \"a1B2c3D4e\"}[/TOOL_RESULTS]"
  },
  {
   "repo": "unsloth/mistral-7b-instruct-v0.3",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/unsloth/mistral-7b-instruct-v0.3/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "e16746b40344d6c5",
   "shape": "B_0131_hidden_tool",
   "args": "str",
   "id": "call_hex24",
   "ok": false,
   "error_type": "TemplateError",
   "error": "Tool call IDs should be alphanumeric strings with length 9!"
  },
  {
   "repo": "unsloth/mistral-7b-instruct-v0.3",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/unsloth/mistral-7b-instruct-v0.3/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "e16746b40344d6c5",
   "shape": "B_0131_hidden_tool",
   "args": "str",
   "id": "uuid4",
   "ok": false,
   "error_type": "TemplateError",
   "error": "Tool call IDs should be alphanumeric strings with length 9!"
  },
  {
   "repo": "unsloth/mistral-7b-instruct-v0.3",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/unsloth/mistral-7b-instruct-v0.3/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "e16746b40344d6c5",
   "shape": "B_0131_hidden_tool",
   "args": "str",
   "id": "alnum9",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "<s>[AVAILABLE_TOOLS] [{\"type\": \"function\", \"function\": {\"name\": \"recall_context\", \"description\": \"Recall saved memories.\", \"parameters\": {\"type\": \"object\", \"properties\": {\"query\": {\"type\": \"string\"}}, \"required\": []}}}][/AVAILABLE_TOOLS][INST] MV2USER hello, what do you remember about me?[/INST][TOOL_CALLS] [{\"name\": \"recall_context\", \"arguments\": \"{\\\"query\\\": \\\"user\\\"}\", \"id\": \"a1B2c3D4e\"}]</s>[TOOL_RESULTS] {\"content\": MV2CTX memory: the user likes short answers, \"call_id\": \"a1B2c3D4e\"}[/TOOL_RESULTS]"
  },
  {
   "repo": "unsloth/mistral-7b-instruct-v0.3",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/unsloth/mistral-7b-instruct-v0.3/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "e16746b40344d6c5",
   "shape": "C_0131_no_system",
   "args": "dict",
   "id": "call_hex24",
   "ok": false,
   "error_type": "TemplateError",
   "error": "Tool call IDs should be alphanumeric strings with length 9!"
  },
  {
   "repo": "unsloth/mistral-7b-instruct-v0.3",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/unsloth/mistral-7b-instruct-v0.3/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "e16746b40344d6c5",
   "shape": "D_0131_content_null",
   "args": "dict",
   "id": "call_hex24",
   "ok": false,
   "error_type": "TemplateError",
   "error": "Tool call IDs should be alphanumeric strings with length 9!"
  },
  {
   "repo": "unsloth/mistral-7b-instruct-v0.3",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/unsloth/mistral-7b-instruct-v0.3/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "e16746b40344d6c5",
   "shape": "E_0131_empty_user",
   "args": "dict",
   "id": "call_hex24",
   "ok": false,
   "error_type": "TemplateError",
   "error": "Tool call IDs should be alphanumeric strings with length 9!"
  },
  {
   "repo": "unsloth/Mistral-Small-3.2-24B-Instruct-2506",
   "fetched": true,
   "file": "chat_template.jinja",
   "url": "https://huggingface.co/unsloth/Mistral-Small-3.2-24B-Instruct-2506/resolve/main/chat_template.jinja",
   "template": "default",
   "sha16": "a126eea7996a0ac9",
   "shape": "A_0130_system_prompt",
   "args": "dict",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "<s>[SYSTEM_PROMPT]You are a helpful assistant.\n\nMV2CTX memory: the user likes short answers[/SYSTEM_PROMPT][AVAILABLE_TOOLS][{\"type\": \"function\", \"function\": {\"name\": \"recall_context\", \"description\": \"Recall saved memories.\", \"parameters\": {\"type\": \"object\", \"properties\": {\"query\": {\"type\": \"string\"}}, \"required\": []}}}][/AVAILABLE_TOOLS][INST]MV2USER hello, what do you remember about me?[/INST]"
  },
  {
   "repo": "unsloth/Mistral-Small-3.2-24B-Instruct-2506",
   "fetched": true,
   "file": "chat_template.jinja",
   "url": "https://huggingface.co/unsloth/Mistral-Small-3.2-24B-Instruct-2506/resolve/main/chat_template.jinja",
   "template": "default",
   "sha16": "a126eea7996a0ac9",
   "shape": "B_0131_hidden_tool",
   "args": "dict",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "<s>[SYSTEM_PROMPT]You are a helpful assistant.[/SYSTEM_PROMPT][AVAILABLE_TOOLS][{\"type\": \"function\", \"function\": {\"name\": \"recall_context\", \"description\": \"Recall saved memories.\", \"parameters\": {\"type\": \"object\", \"properties\": {\"query\": {\"type\": \"string\"}}, \"required\": []}}}][/AVAILABLE_TOOLS][INST]MV2USER hello, what do you remember about me?[/INST][TOOL_CALLS]recall_context[CALL_ID]call_0123456789abcdef01234567[ARGS]{\"query\": \"user\"}</s>[TOOL_RESULTS]call_0123456789abcdef01234567[TOOL_CONTENT]MV2CTX memory: the user likes short answers[/TOOL_RESULTS]"
  },
  {
   "repo": "unsloth/Mistral-Small-3.2-24B-Instruct-2506",
   "fetched": true,
   "file": "chat_template.jinja",
   "url": "https://huggingface.co/unsloth/Mistral-Small-3.2-24B-Instruct-2506/resolve/main/chat_template.jinja",
   "template": "default",
   "sha16": "a126eea7996a0ac9",
   "shape": "B_0131_hidden_tool",
   "args": "dict",
   "id": "uuid4",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "<s>[SYSTEM_PROMPT]You are a helpful assistant.[/SYSTEM_PROMPT][AVAILABLE_TOOLS][{\"type\": \"function\", \"function\": {\"name\": \"recall_context\", \"description\": \"Recall saved memories.\", \"parameters\": {\"type\": \"object\", \"properties\": {\"query\": {\"type\": \"string\"}}, \"required\": []}}}][/AVAILABLE_TOOLS][INST]MV2USER hello, what do you remember about me?[/INST][TOOL_CALLS]recall_context[CALL_ID]3f2b8c1e-9a47-4d2e-b6f1-0c5d7e8a9b10[ARGS]{\"query\": \"user\"}</s>[TOOL_RESULTS]3f2b8c1e-9a47-4d2e-b6f1-0c5d7e8a9b10[TOOL_CONTENT]MV2CTX memory: the user likes short answers[/TOOL_RESULTS]"
  },
  {
   "repo": "unsloth/Mistral-Small-3.2-24B-Instruct-2506",
   "fetched": true,
   "file": "chat_template.jinja",
   "url": "https://huggingface.co/unsloth/Mistral-Small-3.2-24B-Instruct-2506/resolve/main/chat_template.jinja",
   "template": "default",
   "sha16": "a126eea7996a0ac9",
   "shape": "B_0131_hidden_tool",
   "args": "dict",
   "id": "alnum9",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "<s>[SYSTEM_PROMPT]You are a helpful assistant.[/SYSTEM_PROMPT][AVAILABLE_TOOLS][{\"type\": \"function\", \"function\": {\"name\": \"recall_context\", \"description\": \"Recall saved memories.\", \"parameters\": {\"type\": \"object\", \"properties\": {\"query\": {\"type\": \"string\"}}, \"required\": []}}}][/AVAILABLE_TOOLS][INST]MV2USER hello, what do you remember about me?[/INST][TOOL_CALLS]recall_context[CALL_ID]a1B2c3D4e[ARGS]{\"query\": \"user\"}</s>[TOOL_RESULTS]a1B2c3D4e[TOOL_CONTENT]MV2CTX memory: the user likes short answers[/TOOL_RESULTS]"
  },
  {
   "repo": "unsloth/Mistral-Small-3.2-24B-Instruct-2506",
   "fetched": true,
   "file": "chat_template.jinja",
   "url": "https://huggingface.co/unsloth/Mistral-Small-3.2-24B-Instruct-2506/resolve/main/chat_template.jinja",
   "template": "default",
   "sha16": "a126eea7996a0ac9",
   "shape": "B_0131_hidden_tool",
   "args": "str",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "<s>[SYSTEM_PROMPT]You are a helpful assistant.[/SYSTEM_PROMPT][AVAILABLE_TOOLS][{\"type\": \"function\", \"function\": {\"name\": \"recall_context\", \"description\": \"Recall saved memories.\", \"parameters\": {\"type\": \"object\", \"properties\": {\"query\": {\"type\": \"string\"}}, \"required\": []}}}][/AVAILABLE_TOOLS][INST]MV2USER hello, what do you remember about me?[/INST][TOOL_CALLS]recall_context[CALL_ID]call_0123456789abcdef01234567[ARGS]{\"query\": \"user\"}</s>[TOOL_RESULTS]call_0123456789abcdef01234567[TOOL_CONTENT]MV2CTX memory: the user likes short answers[/TOOL_RESULTS]"
  },
  {
   "repo": "unsloth/Mistral-Small-3.2-24B-Instruct-2506",
   "fetched": true,
   "file": "chat_template.jinja",
   "url": "https://huggingface.co/unsloth/Mistral-Small-3.2-24B-Instruct-2506/resolve/main/chat_template.jinja",
   "template": "default",
   "sha16": "a126eea7996a0ac9",
   "shape": "B_0131_hidden_tool",
   "args": "str",
   "id": "uuid4",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "<s>[SYSTEM_PROMPT]You are a helpful assistant.[/SYSTEM_PROMPT][AVAILABLE_TOOLS][{\"type\": \"function\", \"function\": {\"name\": \"recall_context\", \"description\": \"Recall saved memories.\", \"parameters\": {\"type\": \"object\", \"properties\": {\"query\": {\"type\": \"string\"}}, \"required\": []}}}][/AVAILABLE_TOOLS][INST]MV2USER hello, what do you remember about me?[/INST][TOOL_CALLS]recall_context[CALL_ID]3f2b8c1e-9a47-4d2e-b6f1-0c5d7e8a9b10[ARGS]{\"query\": \"user\"}</s>[TOOL_RESULTS]3f2b8c1e-9a47-4d2e-b6f1-0c5d7e8a9b10[TOOL_CONTENT]MV2CTX memory: the user likes short answers[/TOOL_RESULTS]"
  },
  {
   "repo": "unsloth/Mistral-Small-3.2-24B-Instruct-2506",
   "fetched": true,
   "file": "chat_template.jinja",
   "url": "https://huggingface.co/unsloth/Mistral-Small-3.2-24B-Instruct-2506/resolve/main/chat_template.jinja",
   "template": "default",
   "sha16": "a126eea7996a0ac9",
   "shape": "B_0131_hidden_tool",
   "args": "str",
   "id": "alnum9",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "<s>[SYSTEM_PROMPT]You are a helpful assistant.[/SYSTEM_PROMPT][AVAILABLE_TOOLS][{\"type\": \"function\", \"function\": {\"name\": \"recall_context\", \"description\": \"Recall saved memories.\", \"parameters\": {\"type\": \"object\", \"properties\": {\"query\": {\"type\": \"string\"}}, \"required\": []}}}][/AVAILABLE_TOOLS][INST]MV2USER hello, what do you remember about me?[/INST][TOOL_CALLS]recall_context[CALL_ID]a1B2c3D4e[ARGS]{\"query\": \"user\"}</s>[TOOL_RESULTS]a1B2c3D4e[TOOL_CONTENT]MV2CTX memory: the user likes short answers[/TOOL_RESULTS]"
  },
  {
   "repo": "unsloth/Mistral-Small-3.2-24B-Instruct-2506",
   "fetched": true,
   "file": "chat_template.jinja",
   "url": "https://huggingface.co/unsloth/Mistral-Small-3.2-24B-Instruct-2506/resolve/main/chat_template.jinja",
   "template": "default",
   "sha16": "a126eea7996a0ac9",
   "shape": "C_0131_no_system",
   "args": "dict",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": ".\n\nAlways prioritize using tools to provide the most accurate and helpful response. If tools are not available, inform the user that you cannot perform the requested action at the moment.[/SYSTEM_PROMPT][AVAILABLE_TOOLS][{\"type\": \"function\", \"function\": {\"name\": \"recall_context\", \"description\": \"Recall saved memories.\", \"parameters\": {\"type\": \"object\", \"properties\": {\"query\": {\"type\": \"string\"}}, \"required\": []}}}][/AVAILABLE_TOOLS][INST]MV2USER hello, what do you remember about me?[/INST][TOOL_CALLS]recall_context[CALL_ID]call_0123456789abcdef01234567[ARGS]{\"query\": \"user\"}</s>[TOOL_RESULTS]call_0123456789abcdef01234567[TOOL_CONTENT]MV2CTX memory: the user likes short answers[/TOOL_RESULTS]"
  },
  {
   "repo": "unsloth/Mistral-Small-3.2-24B-Instruct-2506",
   "fetched": true,
   "file": "chat_template.jinja",
   "url": "https://huggingface.co/unsloth/Mistral-Small-3.2-24B-Instruct-2506/resolve/main/chat_template.jinja",
   "template": "default",
   "sha16": "a126eea7996a0ac9",
   "shape": "D_0131_content_null",
   "args": "dict",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "<s>[SYSTEM_PROMPT]You are a helpful assistant.[/SYSTEM_PROMPT][AVAILABLE_TOOLS][{\"type\": \"function\", \"function\": {\"name\": \"recall_context\", \"description\": \"Recall saved memories.\", \"parameters\": {\"type\": \"object\", \"properties\": {\"query\": {\"type\": \"string\"}}, \"required\": []}}}][/AVAILABLE_TOOLS][INST]MV2USER hello, what do you remember about me?[/INST]None[TOOL_CALLS]recall_context[CALL_ID]call_0123456789abcdef01234567[ARGS]{\"query\": \"user\"}</s>[TOOL_RESULTS]call_0123456789abcdef01234567[TOOL_CONTENT]MV2CTX memory: the user likes short answers[/TOOL_RESULTS]"
  },
  {
   "repo": "unsloth/Mistral-Small-3.2-24B-Instruct-2506",
   "fetched": true,
   "file": "chat_template.jinja",
   "url": "https://huggingface.co/unsloth/Mistral-Small-3.2-24B-Instruct-2506/resolve/main/chat_template.jinja",
   "template": "default",
   "sha16": "a126eea7996a0ac9",
   "shape": "E_0131_empty_user",
   "args": "dict",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": false,
   "ctx_in_prompt": true,
   "tail": "<s>[SYSTEM_PROMPT]You are a helpful assistant.[/SYSTEM_PROMPT][AVAILABLE_TOOLS][{\"type\": \"function\", \"function\": {\"name\": \"recall_context\", \"description\": \"Recall saved memories.\", \"parameters\": {\"type\": \"object\", \"properties\": {\"query\": {\"type\": \"string\"}}, \"required\": []}}}][/AVAILABLE_TOOLS][INST][/INST][TOOL_CALLS]recall_context[CALL_ID]call_0123456789abcdef01234567[ARGS]{\"query\": \"user\"}</s>[TOOL_RESULTS]call_0123456789abcdef01234567[TOOL_CONTENT]MV2CTX memory: the user likes short answers[/TOOL_RESULTS]"
  },
  {
   "repo": "mistralai/Devstral-Small-2507",
   "fetched": false
  },
  {
   "repo": "microsoft/Phi-4-mini-instruct",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/microsoft/Phi-4-mini-instruct/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "febf589225c9728a",
   "shape": "A_0130_system_prompt",
   "args": "dict",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "<|system|>You are a helpful assistant.\n\nMV2CTX memory: the user likes short answers<|end|><|user|>MV2USER hello, what do you remember about me?<|end|><|assistant|>"
  },
  {
   "repo": "microsoft/Phi-4-mini-instruct",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/microsoft/Phi-4-mini-instruct/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "febf589225c9728a",
   "shape": "B_0131_hidden_tool",
   "args": "dict",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "<|system|>You are a helpful assistant.<|end|><|user|>MV2USER hello, what do you remember about me?<|end|><|assistant|><|end|><|tool|>MV2CTX memory: the user likes short answers<|end|><|assistant|>"
  },
  {
   "repo": "microsoft/Phi-4-mini-instruct",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/microsoft/Phi-4-mini-instruct/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "febf589225c9728a",
   "shape": "B_0131_hidden_tool",
   "args": "dict",
   "id": "uuid4",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "<|system|>You are a helpful assistant.<|end|><|user|>MV2USER hello, what do you remember about me?<|end|><|assistant|><|end|><|tool|>MV2CTX memory: the user likes short answers<|end|><|assistant|>"
  },
  {
   "repo": "microsoft/Phi-4-mini-instruct",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/microsoft/Phi-4-mini-instruct/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "febf589225c9728a",
   "shape": "B_0131_hidden_tool",
   "args": "dict",
   "id": "alnum9",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "<|system|>You are a helpful assistant.<|end|><|user|>MV2USER hello, what do you remember about me?<|end|><|assistant|><|end|><|tool|>MV2CTX memory: the user likes short answers<|end|><|assistant|>"
  },
  {
   "repo": "microsoft/Phi-4-mini-instruct",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/microsoft/Phi-4-mini-instruct/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "febf589225c9728a",
   "shape": "B_0131_hidden_tool",
   "args": "str",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "<|system|>You are a helpful assistant.<|end|><|user|>MV2USER hello, what do you remember about me?<|end|><|assistant|><|end|><|tool|>MV2CTX memory: the user likes short answers<|end|><|assistant|>"
  },
  {
   "repo": "microsoft/Phi-4-mini-instruct",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/microsoft/Phi-4-mini-instruct/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "febf589225c9728a",
   "shape": "B_0131_hidden_tool",
   "args": "str",
   "id": "uuid4",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "<|system|>You are a helpful assistant.<|end|><|user|>MV2USER hello, what do you remember about me?<|end|><|assistant|><|end|><|tool|>MV2CTX memory: the user likes short answers<|end|><|assistant|>"
  },
  {
   "repo": "microsoft/Phi-4-mini-instruct",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/microsoft/Phi-4-mini-instruct/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "febf589225c9728a",
   "shape": "B_0131_hidden_tool",
   "args": "str",
   "id": "alnum9",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "<|system|>You are a helpful assistant.<|end|><|user|>MV2USER hello, what do you remember about me?<|end|><|assistant|><|end|><|tool|>MV2CTX memory: the user likes short answers<|end|><|assistant|>"
  },
  {
   "repo": "microsoft/Phi-4-mini-instruct",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/microsoft/Phi-4-mini-instruct/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "febf589225c9728a",
   "shape": "C_0131_no_system",
   "args": "dict",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "<|user|>MV2USER hello, what do you remember about me?<|end|><|assistant|><|end|><|tool|>MV2CTX memory: the user likes short answers<|end|><|assistant|>"
  },
  {
   "repo": "microsoft/Phi-4-mini-instruct",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/microsoft/Phi-4-mini-instruct/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "febf589225c9728a",
   "shape": "D_0131_content_null",
   "args": "dict",
   "id": "call_hex24",
   "ok": false,
   "error_type": "TypeError",
   "error": "can only concatenate str (not \"NoneType\") to str"
  },
  {
   "repo": "microsoft/Phi-4-mini-instruct",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/microsoft/Phi-4-mini-instruct/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "febf589225c9728a",
   "shape": "E_0131_empty_user",
   "args": "dict",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": false,
   "ctx_in_prompt": true,
   "tail": "<|system|>You are a helpful assistant.<|end|><|user|><|end|><|assistant|><|end|><|tool|>MV2CTX memory: the user likes short answers<|end|><|assistant|>"
  },
  {
   "repo": "openai/gpt-oss-20b",
   "fetched": true,
   "file": "chat_template.jinja",
   "url": "https://huggingface.co/openai/gpt-oss-20b/resolve/main/chat_template.jinja",
   "template": "default",
   "sha16": "a4c9919cbbd4acdd",
   "shape": "A_0130_system_prompt",
   "args": "dict",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "<|start|>system<|message|>You are ChatGPT, a large language model trained by OpenAI.\nKnowledge cutoff: 2024-06\nCurrent date: 2026-11-14\n\nReasoning: medium\n\n# Valid channels: analysis, commentary, final. Channel must be included for every message.\nCalls to these tools must go to the commentary channel: 'functions'.<|end|><|start|>developer<|message|># Instructions\n\nYou are a helpful assistant.\n\nMV2CTX memory: the user likes short answers\n\n# Tools\n\n## functions\n\nnamespace functions {\n\n// Recall saved memories.\ntype recall_context = (_: {\nquery?: string,\n}) => any;\n\n} // namespace functions<|end|><|start|>user<|message|>MV2USER hello, what do you remember about me?<|end|><|start|>assistant"
  },
  {
   "repo": "openai/gpt-oss-20b",
   "fetched": true,
   "file": "chat_template.jinja",
   "url": "https://huggingface.co/openai/gpt-oss-20b/resolve/main/chat_template.jinja",
   "template": "default",
   "sha16": "a4c9919cbbd4acdd",
   "shape": "B_0131_hidden_tool",
   "args": "dict",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "entary, final. Channel must be included for every message.\nCalls to these tools must go to the commentary channel: 'functions'.<|end|><|start|>developer<|message|># Instructions\n\nYou are a helpful assistant.\n\n# Tools\n\n## functions\n\nnamespace functions {\n\n// Recall saved memories.\ntype recall_context = (_: {\nquery?: string,\n}) => any;\n\n} // namespace functions<|end|><|start|>user<|message|>MV2USER hello, what do you remember about me?<|end|><|start|>assistant to=functions.recall_context<|channel|>commentary json<|message|>{\"query\": \"user\"}<|call|><|start|>functions.recall_context to=assistant<|channel|>commentary<|message|>\"MV2CTX memory: the user likes short answers\"<|end|><|start|>assistant"
  },
  {
   "repo": "openai/gpt-oss-20b",
   "fetched": true,
   "file": "chat_template.jinja",
   "url": "https://huggingface.co/openai/gpt-oss-20b/resolve/main/chat_template.jinja",
   "template": "default",
   "sha16": "a4c9919cbbd4acdd",
   "shape": "B_0131_hidden_tool",
   "args": "dict",
   "id": "uuid4",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "entary, final. Channel must be included for every message.\nCalls to these tools must go to the commentary channel: 'functions'.<|end|><|start|>developer<|message|># Instructions\n\nYou are a helpful assistant.\n\n# Tools\n\n## functions\n\nnamespace functions {\n\n// Recall saved memories.\ntype recall_context = (_: {\nquery?: string,\n}) => any;\n\n} // namespace functions<|end|><|start|>user<|message|>MV2USER hello, what do you remember about me?<|end|><|start|>assistant to=functions.recall_context<|channel|>commentary json<|message|>{\"query\": \"user\"}<|call|><|start|>functions.recall_context to=assistant<|channel|>commentary<|message|>\"MV2CTX memory: the user likes short answers\"<|end|><|start|>assistant"
  },
  {
   "repo": "openai/gpt-oss-20b",
   "fetched": true,
   "file": "chat_template.jinja",
   "url": "https://huggingface.co/openai/gpt-oss-20b/resolve/main/chat_template.jinja",
   "template": "default",
   "sha16": "a4c9919cbbd4acdd",
   "shape": "B_0131_hidden_tool",
   "args": "dict",
   "id": "alnum9",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "entary, final. Channel must be included for every message.\nCalls to these tools must go to the commentary channel: 'functions'.<|end|><|start|>developer<|message|># Instructions\n\nYou are a helpful assistant.\n\n# Tools\n\n## functions\n\nnamespace functions {\n\n// Recall saved memories.\ntype recall_context = (_: {\nquery?: string,\n}) => any;\n\n} // namespace functions<|end|><|start|>user<|message|>MV2USER hello, what do you remember about me?<|end|><|start|>assistant to=functions.recall_context<|channel|>commentary json<|message|>{\"query\": \"user\"}<|call|><|start|>functions.recall_context to=assistant<|channel|>commentary<|message|>\"MV2CTX memory: the user likes short answers\"<|end|><|start|>assistant"
  },
  {
   "repo": "openai/gpt-oss-20b",
   "fetched": true,
   "file": "chat_template.jinja",
   "url": "https://huggingface.co/openai/gpt-oss-20b/resolve/main/chat_template.jinja",
   "template": "default",
   "sha16": "a4c9919cbbd4acdd",
   "shape": "B_0131_hidden_tool",
   "args": "str",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": ", final. Channel must be included for every message.\nCalls to these tools must go to the commentary channel: 'functions'.<|end|><|start|>developer<|message|># Instructions\n\nYou are a helpful assistant.\n\n# Tools\n\n## functions\n\nnamespace functions {\n\n// Recall saved memories.\ntype recall_context = (_: {\nquery?: string,\n}) => any;\n\n} // namespace functions<|end|><|start|>user<|message|>MV2USER hello, what do you remember about me?<|end|><|start|>assistant to=functions.recall_context<|channel|>commentary json<|message|>\"{\\\"query\\\": \\\"user\\\"}\"<|call|><|start|>functions.recall_context to=assistant<|channel|>commentary<|message|>\"MV2CTX memory: the user likes short answers\"<|end|><|start|>assistant"
  },
  {
   "repo": "openai/gpt-oss-20b",
   "fetched": true,
   "file": "chat_template.jinja",
   "url": "https://huggingface.co/openai/gpt-oss-20b/resolve/main/chat_template.jinja",
   "template": "default",
   "sha16": "a4c9919cbbd4acdd",
   "shape": "B_0131_hidden_tool",
   "args": "str",
   "id": "uuid4",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": ", final. Channel must be included for every message.\nCalls to these tools must go to the commentary channel: 'functions'.<|end|><|start|>developer<|message|># Instructions\n\nYou are a helpful assistant.\n\n# Tools\n\n## functions\n\nnamespace functions {\n\n// Recall saved memories.\ntype recall_context = (_: {\nquery?: string,\n}) => any;\n\n} // namespace functions<|end|><|start|>user<|message|>MV2USER hello, what do you remember about me?<|end|><|start|>assistant to=functions.recall_context<|channel|>commentary json<|message|>\"{\\\"query\\\": \\\"user\\\"}\"<|call|><|start|>functions.recall_context to=assistant<|channel|>commentary<|message|>\"MV2CTX memory: the user likes short answers\"<|end|><|start|>assistant"
  },
  {
   "repo": "openai/gpt-oss-20b",
   "fetched": true,
   "file": "chat_template.jinja",
   "url": "https://huggingface.co/openai/gpt-oss-20b/resolve/main/chat_template.jinja",
   "template": "default",
   "sha16": "a4c9919cbbd4acdd",
   "shape": "B_0131_hidden_tool",
   "args": "str",
   "id": "alnum9",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": ", final. Channel must be included for every message.\nCalls to these tools must go to the commentary channel: 'functions'.<|end|><|start|>developer<|message|># Instructions\n\nYou are a helpful assistant.\n\n# Tools\n\n## functions\n\nnamespace functions {\n\n// Recall saved memories.\ntype recall_context = (_: {\nquery?: string,\n}) => any;\n\n} // namespace functions<|end|><|start|>user<|message|>MV2USER hello, what do you remember about me?<|end|><|start|>assistant to=functions.recall_context<|channel|>commentary json<|message|>\"{\\\"query\\\": \\\"user\\\"}\"<|call|><|start|>functions.recall_context to=assistant<|channel|>commentary<|message|>\"MV2CTX memory: the user likes short answers\"<|end|><|start|>assistant"
  },
  {
   "repo": "openai/gpt-oss-20b",
   "fetched": true,
   "file": "chat_template.jinja",
   "url": "https://huggingface.co/openai/gpt-oss-20b/resolve/main/chat_template.jinja",
   "template": "default",
   "sha16": "a4c9919cbbd4acdd",
   "shape": "C_0131_no_system",
   "args": "dict",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "ning: medium\n\n# Valid channels: analysis, commentary, final. Channel must be included for every message.\nCalls to these tools must go to the commentary channel: 'functions'.<|end|><|start|>developer<|message|># Tools\n\n## functions\n\nnamespace functions {\n\n// Recall saved memories.\ntype recall_context = (_: {\nquery?: string,\n}) => any;\n\n} // namespace functions<|end|><|start|>user<|message|>MV2USER hello, what do you remember about me?<|end|><|start|>assistant to=functions.recall_context<|channel|>commentary json<|message|>{\"query\": \"user\"}<|call|><|start|>functions.recall_context to=assistant<|channel|>commentary<|message|>\"MV2CTX memory: the user likes short answers\"<|end|><|start|>assistant"
  },
  {
   "repo": "openai/gpt-oss-20b",
   "fetched": true,
   "file": "chat_template.jinja",
   "url": "https://huggingface.co/openai/gpt-oss-20b/resolve/main/chat_template.jinja",
   "template": "default",
   "sha16": "a4c9919cbbd4acdd",
   "shape": "D_0131_content_null",
   "args": "dict",
   "id": "call_hex24",
   "ok": false,
   "error_type": "TypeError",
   "error": "argument of type 'NoneType' is not iterable"
  },
  {
   "repo": "openai/gpt-oss-20b",
   "fetched": true,
   "file": "chat_template.jinja",
   "url": "https://huggingface.co/openai/gpt-oss-20b/resolve/main/chat_template.jinja",
   "template": "default",
   "sha16": "a4c9919cbbd4acdd",
   "shape": "E_0131_empty_user",
   "args": "dict",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": false,
   "ctx_in_prompt": true,
   "tail": "ing: medium\n\n# Valid channels: analysis, commentary, final. Channel must be included for every message.\nCalls to these tools must go to the commentary channel: 'functions'.<|end|><|start|>developer<|message|># Instructions\n\nYou are a helpful assistant.\n\n# Tools\n\n## functions\n\nnamespace functions {\n\n// Recall saved memories.\ntype recall_context = (_: {\nquery?: string,\n}) => any;\n\n} // namespace functions<|end|><|start|>user<|message|><|end|><|start|>assistant to=functions.recall_context<|channel|>commentary json<|message|>{\"query\": \"user\"}<|call|><|start|>functions.recall_context to=assistant<|channel|>commentary<|message|>\"MV2CTX memory: the user likes short answers\"<|end|><|start|>assistant"
  },
  {
   "repo": "zai-org/GLM-4.5-Air",
   "fetched": true,
   "file": "chat_template.jinja",
   "url": "https://huggingface.co/zai-org/GLM-4.5-Air/resolve/main/chat_template.jinja",
   "template": "default",
   "sha16": "44f815868bf02fa4",
   "shape": "A_0130_system_prompt",
   "args": "dict",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": " with function signatures within <tools></tools> XML tags:\n<tools>\n{\"type\": \"function\", \"function\": {\"name\": \"recall_context\", \"description\": \"Recall saved memories.\", \"parameters\": {\"type\": \"object\", \"properties\": {\"query\": {\"type\": \"string\"}}, \"required\": []}}}\n</tools>\n\nFor each function call, output the function name and arguments within the following XML format:\n<tool_call>{function-name}\n<arg_key>{arg-key-1}</arg_key>\n<arg_value>{arg-value-1}</arg_value>\n<arg_key>{arg-key-2}</arg_key>\n<arg_value>{arg-value-2}</arg_value>\n...\n</tool_call><|system|>\nYou are a helpful assistant.\n\nMV2CTX memory: the user likes short answers<|user|>\nMV2USER hello, what do you remember about me?<|assistant|>"
  },
  {
   "repo": "zai-org/GLM-4.5-Air",
   "fetched": true,
   "file": "chat_template.jinja",
   "url": "https://huggingface.co/zai-org/GLM-4.5-Air/resolve/main/chat_template.jinja",
   "template": "default",
   "sha16": "44f815868bf02fa4",
   "shape": "B_0131_hidden_tool",
   "args": "dict",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "\"parameters\": {\"type\": \"object\", \"properties\": {\"query\": {\"type\": \"string\"}}, \"required\": []}}}\n</tools>\n\nFor each function call, output the function name and arguments within the following XML format:\n<tool_call>{function-name}\n<arg_key>{arg-key-1}</arg_key>\n<arg_value>{arg-value-1}</arg_value>\n<arg_key>{arg-key-2}</arg_key>\n<arg_value>{arg-value-2}</arg_value>\n...\n</tool_call><|system|>\nYou are a helpful assistant.<|user|>\nMV2USER hello, what do you remember about me?<|assistant|>\n<think></think>\n<tool_call>recall_context\n<arg_key>query</arg_key>\n<arg_value>user</arg_value>\n</tool_call><|observation|>\n<tool_response>\nMV2CTX memory: the user likes short answers\n</tool_response><|assistant|>"
  },
  {
   "repo": "zai-org/GLM-4.5-Air",
   "fetched": true,
   "file": "chat_template.jinja",
   "url": "https://huggingface.co/zai-org/GLM-4.5-Air/resolve/main/chat_template.jinja",
   "template": "default",
   "sha16": "44f815868bf02fa4",
   "shape": "B_0131_hidden_tool",
   "args": "dict",
   "id": "uuid4",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "\"parameters\": {\"type\": \"object\", \"properties\": {\"query\": {\"type\": \"string\"}}, \"required\": []}}}\n</tools>\n\nFor each function call, output the function name and arguments within the following XML format:\n<tool_call>{function-name}\n<arg_key>{arg-key-1}</arg_key>\n<arg_value>{arg-value-1}</arg_value>\n<arg_key>{arg-key-2}</arg_key>\n<arg_value>{arg-value-2}</arg_value>\n...\n</tool_call><|system|>\nYou are a helpful assistant.<|user|>\nMV2USER hello, what do you remember about me?<|assistant|>\n<think></think>\n<tool_call>recall_context\n<arg_key>query</arg_key>\n<arg_value>user</arg_value>\n</tool_call><|observation|>\n<tool_response>\nMV2CTX memory: the user likes short answers\n</tool_response><|assistant|>"
  },
  {
   "repo": "zai-org/GLM-4.5-Air",
   "fetched": true,
   "file": "chat_template.jinja",
   "url": "https://huggingface.co/zai-org/GLM-4.5-Air/resolve/main/chat_template.jinja",
   "template": "default",
   "sha16": "44f815868bf02fa4",
   "shape": "B_0131_hidden_tool",
   "args": "dict",
   "id": "alnum9",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "\"parameters\": {\"type\": \"object\", \"properties\": {\"query\": {\"type\": \"string\"}}, \"required\": []}}}\n</tools>\n\nFor each function call, output the function name and arguments within the following XML format:\n<tool_call>{function-name}\n<arg_key>{arg-key-1}</arg_key>\n<arg_value>{arg-value-1}</arg_value>\n<arg_key>{arg-key-2}</arg_key>\n<arg_value>{arg-value-2}</arg_value>\n...\n</tool_call><|system|>\nYou are a helpful assistant.<|user|>\nMV2USER hello, what do you remember about me?<|assistant|>\n<think></think>\n<tool_call>recall_context\n<arg_key>query</arg_key>\n<arg_value>user</arg_value>\n</tool_call><|observation|>\n<tool_response>\nMV2CTX memory: the user likes short answers\n</tool_response><|assistant|>"
  },
  {
   "repo": "zai-org/GLM-4.5-Air",
   "fetched": true,
   "file": "chat_template.jinja",
   "url": "https://huggingface.co/zai-org/GLM-4.5-Air/resolve/main/chat_template.jinja",
   "template": "default",
   "sha16": "44f815868bf02fa4",
   "shape": "B_0131_hidden_tool",
   "args": "str",
   "id": "call_hex24",
   "ok": false,
   "error_type": "UndefinedError",
   "error": "'str object' has no attribute 'items'"
  },
  {
   "repo": "zai-org/GLM-4.5-Air",
   "fetched": true,
   "file": "chat_template.jinja",
   "url": "https://huggingface.co/zai-org/GLM-4.5-Air/resolve/main/chat_template.jinja",
   "template": "default",
   "sha16": "44f815868bf02fa4",
   "shape": "B_0131_hidden_tool",
   "args": "str",
   "id": "uuid4",
   "ok": false,
   "error_type": "UndefinedError",
   "error": "'str object' has no attribute 'items'"
  },
  {
   "repo": "zai-org/GLM-4.5-Air",
   "fetched": true,
   "file": "chat_template.jinja",
   "url": "https://huggingface.co/zai-org/GLM-4.5-Air/resolve/main/chat_template.jinja",
   "template": "default",
   "sha16": "44f815868bf02fa4",
   "shape": "B_0131_hidden_tool",
   "args": "str",
   "id": "alnum9",
   "ok": false,
   "error_type": "UndefinedError",
   "error": "'str object' has no attribute 'items'"
  },
  {
   "repo": "zai-org/GLM-4.5-Air",
   "fetched": true,
   "file": "chat_template.jinja",
   "url": "https://huggingface.co/zai-org/GLM-4.5-Air/resolve/main/chat_template.jinja",
   "template": "default",
   "sha16": "44f815868bf02fa4",
   "shape": "C_0131_no_system",
   "args": "dict",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "escription\": \"Recall saved memories.\", \"parameters\": {\"type\": \"object\", \"properties\": {\"query\": {\"type\": \"string\"}}, \"required\": []}}}\n</tools>\n\nFor each function call, output the function name and arguments within the following XML format:\n<tool_call>{function-name}\n<arg_key>{arg-key-1}</arg_key>\n<arg_value>{arg-value-1}</arg_value>\n<arg_key>{arg-key-2}</arg_key>\n<arg_value>{arg-value-2}</arg_value>\n...\n</tool_call><|user|>\nMV2USER hello, what do you remember about me?<|assistant|>\n<think></think>\n<tool_call>recall_context\n<arg_key>query</arg_key>\n<arg_value>user</arg_value>\n</tool_call><|observation|>\n<tool_response>\nMV2CTX memory: the user likes short answers\n</tool_response><|assistant|>"
  },
  {
   "repo": "zai-org/GLM-4.5-Air",
   "fetched": true,
   "file": "chat_template.jinja",
   "url": "https://huggingface.co/zai-org/GLM-4.5-Air/resolve/main/chat_template.jinja",
   "template": "default",
   "sha16": "44f815868bf02fa4",
   "shape": "D_0131_content_null",
   "args": "dict",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "meters\": {\"type\": \"object\", \"properties\": {\"query\": {\"type\": \"string\"}}, \"required\": []}}}\n</tools>\n\nFor each function call, output the function name and arguments within the following XML format:\n<tool_call>{function-name}\n<arg_key>{arg-key-1}</arg_key>\n<arg_value>{arg-value-1}</arg_value>\n<arg_key>{arg-key-2}</arg_key>\n<arg_value>{arg-value-2}</arg_value>\n...\n</tool_call><|system|>\nYou are a helpful assistant.<|user|>\nMV2USER hello, what do you remember about me?<|assistant|>\n<think></think>\nNone\n<tool_call>recall_context\n<arg_key>query</arg_key>\n<arg_value>user</arg_value>\n</tool_call><|observation|>\n<tool_response>\nMV2CTX memory: the user likes short answers\n</tool_response><|assistant|>"
  },
  {
   "repo": "zai-org/GLM-4.5-Air",
   "fetched": true,
   "file": "chat_template.jinja",
   "url": "https://huggingface.co/zai-org/GLM-4.5-Air/resolve/main/chat_template.jinja",
   "template": "default",
   "sha16": "44f815868bf02fa4",
   "shape": "E_0131_empty_user",
   "args": "dict",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": false,
   "ctx_in_prompt": true,
   "tail": "t\", \"description\": \"Recall saved memories.\", \"parameters\": {\"type\": \"object\", \"properties\": {\"query\": {\"type\": \"string\"}}, \"required\": []}}}\n</tools>\n\nFor each function call, output the function name and arguments within the following XML format:\n<tool_call>{function-name}\n<arg_key>{arg-key-1}</arg_key>\n<arg_value>{arg-value-1}</arg_value>\n<arg_key>{arg-key-2}</arg_key>\n<arg_value>{arg-value-2}</arg_value>\n...\n</tool_call><|system|>\nYou are a helpful assistant.<|user|>\n<|assistant|>\n<think></think>\n<tool_call>recall_context\n<arg_key>query</arg_key>\n<arg_value>user</arg_value>\n</tool_call><|observation|>\n<tool_response>\nMV2CTX memory: the user likes short answers\n</tool_response><|assistant|>"
  },
  {
   "repo": "NousResearch/Hermes-3-Llama-3.1-8B",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/NousResearch/Hermes-3-Llama-3.1-8B/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "9ff2eb30a1d3e937",
   "shape": "A_0130_system_prompt",
   "args": "dict",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "<s><|im_start|>system\nYou are a helpful assistant.\n\nMV2CTX memory: the user likes short answers<|im_end|>\n<|im_start|>user\nMV2USER hello, what do you remember about me?<|im_end|>\n<|im_start|>assistant\n"
  },
  {
   "repo": "NousResearch/Hermes-3-Llama-3.1-8B",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/NousResearch/Hermes-3-Llama-3.1-8B/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "9ff2eb30a1d3e937",
   "shape": "B_0131_hidden_tool",
   "args": "dict",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "<s><|im_start|>system\nYou are a helpful assistant.<|im_end|>\n<|im_start|>user\nMV2USER hello, what do you remember about me?<|im_end|>\n<|im_start|>assistant\n<|im_end|>\n<|im_start|>tool\nMV2CTX memory: the user likes short answers<|im_end|>\n<|im_start|>assistant\n"
  },
  {
   "repo": "NousResearch/Hermes-3-Llama-3.1-8B",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/NousResearch/Hermes-3-Llama-3.1-8B/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "9ff2eb30a1d3e937",
   "shape": "B_0131_hidden_tool",
   "args": "dict",
   "id": "uuid4",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "<s><|im_start|>system\nYou are a helpful assistant.<|im_end|>\n<|im_start|>user\nMV2USER hello, what do you remember about me?<|im_end|>\n<|im_start|>assistant\n<|im_end|>\n<|im_start|>tool\nMV2CTX memory: the user likes short answers<|im_end|>\n<|im_start|>assistant\n"
  },
  {
   "repo": "NousResearch/Hermes-3-Llama-3.1-8B",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/NousResearch/Hermes-3-Llama-3.1-8B/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "9ff2eb30a1d3e937",
   "shape": "B_0131_hidden_tool",
   "args": "dict",
   "id": "alnum9",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "<s><|im_start|>system\nYou are a helpful assistant.<|im_end|>\n<|im_start|>user\nMV2USER hello, what do you remember about me?<|im_end|>\n<|im_start|>assistant\n<|im_end|>\n<|im_start|>tool\nMV2CTX memory: the user likes short answers<|im_end|>\n<|im_start|>assistant\n"
  },
  {
   "repo": "NousResearch/Hermes-3-Llama-3.1-8B",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/NousResearch/Hermes-3-Llama-3.1-8B/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "9ff2eb30a1d3e937",
   "shape": "B_0131_hidden_tool",
   "args": "str",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "<s><|im_start|>system\nYou are a helpful assistant.<|im_end|>\n<|im_start|>user\nMV2USER hello, what do you remember about me?<|im_end|>\n<|im_start|>assistant\n<|im_end|>\n<|im_start|>tool\nMV2CTX memory: the user likes short answers<|im_end|>\n<|im_start|>assistant\n"
  },
  {
   "repo": "NousResearch/Hermes-3-Llama-3.1-8B",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/NousResearch/Hermes-3-Llama-3.1-8B/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "9ff2eb30a1d3e937",
   "shape": "B_0131_hidden_tool",
   "args": "str",
   "id": "uuid4",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "<s><|im_start|>system\nYou are a helpful assistant.<|im_end|>\n<|im_start|>user\nMV2USER hello, what do you remember about me?<|im_end|>\n<|im_start|>assistant\n<|im_end|>\n<|im_start|>tool\nMV2CTX memory: the user likes short answers<|im_end|>\n<|im_start|>assistant\n"
  },
  {
   "repo": "NousResearch/Hermes-3-Llama-3.1-8B",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/NousResearch/Hermes-3-Llama-3.1-8B/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "9ff2eb30a1d3e937",
   "shape": "B_0131_hidden_tool",
   "args": "str",
   "id": "alnum9",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "<s><|im_start|>system\nYou are a helpful assistant.<|im_end|>\n<|im_start|>user\nMV2USER hello, what do you remember about me?<|im_end|>\n<|im_start|>assistant\n<|im_end|>\n<|im_start|>tool\nMV2CTX memory: the user likes short answers<|im_end|>\n<|im_start|>assistant\n"
  },
  {
   "repo": "NousResearch/Hermes-3-Llama-3.1-8B",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/NousResearch/Hermes-3-Llama-3.1-8B/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "9ff2eb30a1d3e937",
   "shape": "C_0131_no_system",
   "args": "dict",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "<s><|im_start|>system\nYou are a helpful assistant.<|im_end|>\n<|im_start|>user\nMV2USER hello, what do you remember about me?<|im_end|>\n<|im_start|>assistant\n<|im_end|>\n<|im_start|>tool\nMV2CTX memory: the user likes short answers<|im_end|>\n<|im_start|>assistant\n"
  },
  {
   "repo": "NousResearch/Hermes-3-Llama-3.1-8B",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/NousResearch/Hermes-3-Llama-3.1-8B/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "9ff2eb30a1d3e937",
   "shape": "D_0131_content_null",
   "args": "dict",
   "id": "call_hex24",
   "ok": false,
   "error_type": "TypeError",
   "error": "can only concatenate str (not \"NoneType\") to str"
  },
  {
   "repo": "NousResearch/Hermes-3-Llama-3.1-8B",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/NousResearch/Hermes-3-Llama-3.1-8B/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "9ff2eb30a1d3e937",
   "shape": "E_0131_empty_user",
   "args": "dict",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": false,
   "ctx_in_prompt": true,
   "tail": "<s><|im_start|>system\nYou are a helpful assistant.<|im_end|>\n<|im_start|>user\n<|im_end|>\n<|im_start|>assistant\n<|im_end|>\n<|im_start|>tool\nMV2CTX memory: the user likes short answers<|im_end|>\n<|im_start|>assistant\n"
  },
  {
   "repo": "NousResearch/Hermes-3-Llama-3.1-8B",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/NousResearch/Hermes-3-Llama-3.1-8B/resolve/main/tokenizer_config.json",
   "template": "tool_use",
   "sha16": "7ce09d55d3690e06",
   "shape": "A_0130_system_prompt",
   "args": "dict",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "}} </tools>Use the following pydantic model json schema for each tool call you will make: {\"properties\": {\"name\": {\"title\": \"Name\", \"type\": \"string\"}, \"arguments\": {\"title\": \"Arguments\", \"type\": \"object\"}}, \"required\": [\"name\", \"arguments\"], \"title\": \"FunctionCall\", \"type\": \"object\"}}\nFor each function call return a json object with function name and arguments within <tool_call></tool_call> XML tags as follows:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-dict>}\n</tool_call><|im_end|>\n<|im_start|>system\nYou are a helpful assistant.\n\nMV2CTX memory: the user likes short answers<|im_end|>\n<|im_start|>user\nMV2USER hello, what do you remember about me?<|im_end|>\n<|im_start|>assistant\n"
  },
  {
   "repo": "NousResearch/Hermes-3-Llama-3.1-8B",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/NousResearch/Hermes-3-Llama-3.1-8B/resolve/main/tokenizer_config.json",
   "template": "tool_use",
   "sha16": "7ce09d55d3690e06",
   "shape": "B_0131_hidden_tool",
   "args": "dict",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "\"Arguments\", \"type\": \"object\"}}, \"required\": [\"name\", \"arguments\"], \"title\": \"FunctionCall\", \"type\": \"object\"}}\nFor each function call return a json object with function name and arguments within <tool_call></tool_call> XML tags as follows:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-dict>}\n</tool_call><|im_end|>\n<|im_start|>system\nYou are a helpful assistant.<|im_end|>\n<|im_start|>user\nMV2USER hello, what do you remember about me?<|im_end|>\n<|im_start|>assistant\n<tool_call>\n{\"name\": \"recall_context\", \"arguments\": {\"query\": \"user\"}}\n</tool_call><|im_end|>\n<|im_start|>tool\n<tool_response>\nMV2CTX memory: the user likes short answers\n</tool_response><|im_end|><|im_start|>assistant\n"
  },
  {
   "repo": "NousResearch/Hermes-3-Llama-3.1-8B",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/NousResearch/Hermes-3-Llama-3.1-8B/resolve/main/tokenizer_config.json",
   "template": "tool_use",
   "sha16": "7ce09d55d3690e06",
   "shape": "B_0131_hidden_tool",
   "args": "dict",
   "id": "uuid4",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "\"Arguments\", \"type\": \"object\"}}, \"required\": [\"name\", \"arguments\"], \"title\": \"FunctionCall\", \"type\": \"object\"}}\nFor each function call return a json object with function name and arguments within <tool_call></tool_call> XML tags as follows:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-dict>}\n</tool_call><|im_end|>\n<|im_start|>system\nYou are a helpful assistant.<|im_end|>\n<|im_start|>user\nMV2USER hello, what do you remember about me?<|im_end|>\n<|im_start|>assistant\n<tool_call>\n{\"name\": \"recall_context\", \"arguments\": {\"query\": \"user\"}}\n</tool_call><|im_end|>\n<|im_start|>tool\n<tool_response>\nMV2CTX memory: the user likes short answers\n</tool_response><|im_end|><|im_start|>assistant\n"
  },
  {
   "repo": "NousResearch/Hermes-3-Llama-3.1-8B",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/NousResearch/Hermes-3-Llama-3.1-8B/resolve/main/tokenizer_config.json",
   "template": "tool_use",
   "sha16": "7ce09d55d3690e06",
   "shape": "B_0131_hidden_tool",
   "args": "dict",
   "id": "alnum9",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "\"Arguments\", \"type\": \"object\"}}, \"required\": [\"name\", \"arguments\"], \"title\": \"FunctionCall\", \"type\": \"object\"}}\nFor each function call return a json object with function name and arguments within <tool_call></tool_call> XML tags as follows:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-dict>}\n</tool_call><|im_end|>\n<|im_start|>system\nYou are a helpful assistant.<|im_end|>\n<|im_start|>user\nMV2USER hello, what do you remember about me?<|im_end|>\n<|im_start|>assistant\n<tool_call>\n{\"name\": \"recall_context\", \"arguments\": {\"query\": \"user\"}}\n</tool_call><|im_end|>\n<|im_start|>tool\n<tool_response>\nMV2CTX memory: the user likes short answers\n</tool_response><|im_end|><|im_start|>assistant\n"
  },
  {
   "repo": "NousResearch/Hermes-3-Llama-3.1-8B",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/NousResearch/Hermes-3-Llama-3.1-8B/resolve/main/tokenizer_config.json",
   "template": "tool_use",
   "sha16": "7ce09d55d3690e06",
   "shape": "B_0131_hidden_tool",
   "args": "str",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "\"Arguments\", \"type\": \"object\"}}, \"required\": [\"name\", \"arguments\"], \"title\": \"FunctionCall\", \"type\": \"object\"}}\nFor each function call return a json object with function name and arguments within <tool_call></tool_call> XML tags as follows:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-dict>}\n</tool_call><|im_end|>\n<|im_start|>system\nYou are a helpful assistant.<|im_end|>\n<|im_start|>user\nMV2USER hello, what do you remember about me?<|im_end|>\n<|im_start|>assistant\n<tool_call>\n{\"name\": \"recall_context\", \"arguments\": {\"query\": \"user\"}}\n</tool_call><|im_end|>\n<|im_start|>tool\n<tool_response>\nMV2CTX memory: the user likes short answers\n</tool_response><|im_end|><|im_start|>assistant\n"
  },
  {
   "repo": "NousResearch/Hermes-3-Llama-3.1-8B",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/NousResearch/Hermes-3-Llama-3.1-8B/resolve/main/tokenizer_config.json",
   "template": "tool_use",
   "sha16": "7ce09d55d3690e06",
   "shape": "B_0131_hidden_tool",
   "args": "str",
   "id": "uuid4",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "\"Arguments\", \"type\": \"object\"}}, \"required\": [\"name\", \"arguments\"], \"title\": \"FunctionCall\", \"type\": \"object\"}}\nFor each function call return a json object with function name and arguments within <tool_call></tool_call> XML tags as follows:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-dict>}\n</tool_call><|im_end|>\n<|im_start|>system\nYou are a helpful assistant.<|im_end|>\n<|im_start|>user\nMV2USER hello, what do you remember about me?<|im_end|>\n<|im_start|>assistant\n<tool_call>\n{\"name\": \"recall_context\", \"arguments\": {\"query\": \"user\"}}\n</tool_call><|im_end|>\n<|im_start|>tool\n<tool_response>\nMV2CTX memory: the user likes short answers\n</tool_response><|im_end|><|im_start|>assistant\n"
  },
  {
   "repo": "NousResearch/Hermes-3-Llama-3.1-8B",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/NousResearch/Hermes-3-Llama-3.1-8B/resolve/main/tokenizer_config.json",
   "template": "tool_use",
   "sha16": "7ce09d55d3690e06",
   "shape": "B_0131_hidden_tool",
   "args": "str",
   "id": "alnum9",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "\"Arguments\", \"type\": \"object\"}}, \"required\": [\"name\", \"arguments\"], \"title\": \"FunctionCall\", \"type\": \"object\"}}\nFor each function call return a json object with function name and arguments within <tool_call></tool_call> XML tags as follows:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-dict>}\n</tool_call><|im_end|>\n<|im_start|>system\nYou are a helpful assistant.<|im_end|>\n<|im_start|>user\nMV2USER hello, what do you remember about me?<|im_end|>\n<|im_start|>assistant\n<tool_call>\n{\"name\": \"recall_context\", \"arguments\": {\"query\": \"user\"}}\n</tool_call><|im_end|>\n<|im_start|>tool\n<tool_response>\nMV2CTX memory: the user likes short answers\n</tool_response><|im_end|><|im_start|>assistant\n"
  },
  {
   "repo": "NousResearch/Hermes-3-Llama-3.1-8B",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/NousResearch/Hermes-3-Llama-3.1-8B/resolve/main/tokenizer_config.json",
   "template": "tool_use",
   "sha16": "7ce09d55d3690e06",
   "shape": "C_0131_no_system",
   "args": "dict",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "title\": \"Name\", \"type\": \"string\"}, \"arguments\": {\"title\": \"Arguments\", \"type\": \"object\"}}, \"required\": [\"name\", \"arguments\"], \"title\": \"FunctionCall\", \"type\": \"object\"}}\nFor each function call return a json object with function name and arguments within <tool_call></tool_call> XML tags as follows:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-dict>}\n</tool_call><|im_end|>\n<|im_start|>user\nMV2USER hello, what do you remember about me?<|im_end|>\n<|im_start|>assistant\n<tool_call>\n{\"name\": \"recall_context\", \"arguments\": {\"query\": \"user\"}}\n</tool_call><|im_end|>\n<|im_start|>tool\n<tool_response>\nMV2CTX memory: the user likes short answers\n</tool_response><|im_end|><|im_start|>assistant\n"
  },
  {
   "repo": "NousResearch/Hermes-3-Llama-3.1-8B",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/NousResearch/Hermes-3-Llama-3.1-8B/resolve/main/tokenizer_config.json",
   "template": "tool_use",
   "sha16": "7ce09d55d3690e06",
   "shape": "D_0131_content_null",
   "args": "dict",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "\"Arguments\", \"type\": \"object\"}}, \"required\": [\"name\", \"arguments\"], \"title\": \"FunctionCall\", \"type\": \"object\"}}\nFor each function call return a json object with function name and arguments within <tool_call></tool_call> XML tags as follows:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-dict>}\n</tool_call><|im_end|>\n<|im_start|>system\nYou are a helpful assistant.<|im_end|>\n<|im_start|>user\nMV2USER hello, what do you remember about me?<|im_end|>\n<|im_start|>assistant\n<tool_call>\n{\"name\": \"recall_context\", \"arguments\": {\"query\": \"user\"}}\n</tool_call><|im_end|>\n<|im_start|>tool\n<tool_response>\nMV2CTX memory: the user likes short answers\n</tool_response><|im_end|><|im_start|>assistant\n"
  },
  {
   "repo": "NousResearch/Hermes-3-Llama-3.1-8B",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/NousResearch/Hermes-3-Llama-3.1-8B/resolve/main/tokenizer_config.json",
   "template": "tool_use",
   "sha16": "7ce09d55d3690e06",
   "shape": "E_0131_empty_user",
   "args": "dict",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": false,
   "ctx_in_prompt": true,
   "tail": "\", \"type\": \"string\"}, \"arguments\": {\"title\": \"Arguments\", \"type\": \"object\"}}, \"required\": [\"name\", \"arguments\"], \"title\": \"FunctionCall\", \"type\": \"object\"}}\nFor each function call return a json object with function name and arguments within <tool_call></tool_call> XML tags as follows:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-dict>}\n</tool_call><|im_end|>\n<|im_start|>system\nYou are a helpful assistant.<|im_end|>\n<|im_start|>user\n<|im_end|>\n<|im_start|>assistant\n<tool_call>\n{\"name\": \"recall_context\", \"arguments\": {\"query\": \"user\"}}\n</tool_call><|im_end|>\n<|im_start|>tool\n<tool_response>\nMV2CTX memory: the user likes short answers\n</tool_response><|im_end|><|im_start|>assistant\n"
  },
  {
   "repo": "ibm-granite/granite-3.3-8b-instruct",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/ibm-granite/granite-3.3-8b-instruct/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "0bea7711316005a1",
   "shape": "A_0130_system_prompt",
   "args": "dict",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "a helpful assistant.\n\nMV2CTX memory: the user likes short answers<|end_of_text|>\n<|start_of_role|>available_tools<|end_of_role|>[\n    {\n        \"type\": \"function\",\n        \"function\": {\n            \"name\": \"recall_context\",\n            \"description\": \"Recall saved memories.\",\n            \"parameters\": {\n                \"type\": \"object\",\n                \"properties\": {\n                    \"query\": {\n                        \"type\": \"string\"\n                    }\n                },\n                \"required\": []\n            }\n        }\n    }\n]<|end_of_text|>\n<|start_of_role|>user<|end_of_role|>MV2USER hello, what do you remember about me?<|end_of_text|>\n<|start_of_role|>assistant<|end_of_role|>"
  },
  {
   "repo": "ibm-granite/granite-3.3-8b-instruct",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/ibm-granite/granite-3.3-8b-instruct/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "0bea7711316005a1",
   "shape": "B_0131_hidden_tool",
   "args": "dict",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "\"function\",\n        \"function\": {\n            \"name\": \"recall_context\",\n            \"description\": \"Recall saved memories.\",\n            \"parameters\": {\n                \"type\": \"object\",\n                \"properties\": {\n                    \"query\": {\n                        \"type\": \"string\"\n                    }\n                },\n                \"required\": []\n            }\n        }\n    }\n]<|end_of_text|>\n<|start_of_role|>user<|end_of_role|>MV2USER hello, what do you remember about me?<|end_of_text|>\n<|start_of_role|>assistant<|end_of_role|><|end_of_text|>\n<|start_of_role|>tool<|end_of_role|>MV2CTX memory: the user likes short answers<|end_of_text|>\n<|start_of_role|>assistant<|end_of_role|>"
  },
  {
   "repo": "ibm-granite/granite-3.3-8b-instruct",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/ibm-granite/granite-3.3-8b-instruct/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "0bea7711316005a1",
   "shape": "B_0131_hidden_tool",
   "args": "dict",
   "id": "uuid4",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "\"function\",\n        \"function\": {\n            \"name\": \"recall_context\",\n            \"description\": \"Recall saved memories.\",\n            \"parameters\": {\n                \"type\": \"object\",\n                \"properties\": {\n                    \"query\": {\n                        \"type\": \"string\"\n                    }\n                },\n                \"required\": []\n            }\n        }\n    }\n]<|end_of_text|>\n<|start_of_role|>user<|end_of_role|>MV2USER hello, what do you remember about me?<|end_of_text|>\n<|start_of_role|>assistant<|end_of_role|><|end_of_text|>\n<|start_of_role|>tool<|end_of_role|>MV2CTX memory: the user likes short answers<|end_of_text|>\n<|start_of_role|>assistant<|end_of_role|>"
  },
  {
   "repo": "ibm-granite/granite-3.3-8b-instruct",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/ibm-granite/granite-3.3-8b-instruct/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "0bea7711316005a1",
   "shape": "B_0131_hidden_tool",
   "args": "dict",
   "id": "alnum9",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "\"function\",\n        \"function\": {\n            \"name\": \"recall_context\",\n            \"description\": \"Recall saved memories.\",\n            \"parameters\": {\n                \"type\": \"object\",\n                \"properties\": {\n                    \"query\": {\n                        \"type\": \"string\"\n                    }\n                },\n                \"required\": []\n            }\n        }\n    }\n]<|end_of_text|>\n<|start_of_role|>user<|end_of_role|>MV2USER hello, what do you remember about me?<|end_of_text|>\n<|start_of_role|>assistant<|end_of_role|><|end_of_text|>\n<|start_of_role|>tool<|end_of_role|>MV2CTX memory: the user likes short answers<|end_of_text|>\n<|start_of_role|>assistant<|end_of_role|>"
  },
  {
   "repo": "ibm-granite/granite-3.3-8b-instruct",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/ibm-granite/granite-3.3-8b-instruct/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "0bea7711316005a1",
   "shape": "B_0131_hidden_tool",
   "args": "str",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "\"function\",\n        \"function\": {\n            \"name\": \"recall_context\",\n            \"description\": \"Recall saved memories.\",\n            \"parameters\": {\n                \"type\": \"object\",\n                \"properties\": {\n                    \"query\": {\n                        \"type\": \"string\"\n                    }\n                },\n                \"required\": []\n            }\n        }\n    }\n]<|end_of_text|>\n<|start_of_role|>user<|end_of_role|>MV2USER hello, what do you remember about me?<|end_of_text|>\n<|start_of_role|>assistant<|end_of_role|><|end_of_text|>\n<|start_of_role|>tool<|end_of_role|>MV2CTX memory: the user likes short answers<|end_of_text|>\n<|start_of_role|>assistant<|end_of_role|>"
  },
  {
   "repo": "ibm-granite/granite-3.3-8b-instruct",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/ibm-granite/granite-3.3-8b-instruct/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "0bea7711316005a1",
   "shape": "B_0131_hidden_tool",
   "args": "str",
   "id": "uuid4",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "\"function\",\n        \"function\": {\n            \"name\": \"recall_context\",\n            \"description\": \"Recall saved memories.\",\n            \"parameters\": {\n                \"type\": \"object\",\n                \"properties\": {\n                    \"query\": {\n                        \"type\": \"string\"\n                    }\n                },\n                \"required\": []\n            }\n        }\n    }\n]<|end_of_text|>\n<|start_of_role|>user<|end_of_role|>MV2USER hello, what do you remember about me?<|end_of_text|>\n<|start_of_role|>assistant<|end_of_role|><|end_of_text|>\n<|start_of_role|>tool<|end_of_role|>MV2CTX memory: the user likes short answers<|end_of_text|>\n<|start_of_role|>assistant<|end_of_role|>"
  },
  {
   "repo": "ibm-granite/granite-3.3-8b-instruct",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/ibm-granite/granite-3.3-8b-instruct/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "0bea7711316005a1",
   "shape": "B_0131_hidden_tool",
   "args": "str",
   "id": "alnum9",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "\"function\",\n        \"function\": {\n            \"name\": \"recall_context\",\n            \"description\": \"Recall saved memories.\",\n            \"parameters\": {\n                \"type\": \"object\",\n                \"properties\": {\n                    \"query\": {\n                        \"type\": \"string\"\n                    }\n                },\n                \"required\": []\n            }\n        }\n    }\n]<|end_of_text|>\n<|start_of_role|>user<|end_of_role|>MV2USER hello, what do you remember about me?<|end_of_text|>\n<|start_of_role|>assistant<|end_of_role|><|end_of_text|>\n<|start_of_role|>tool<|end_of_role|>MV2CTX memory: the user likes short answers<|end_of_text|>\n<|start_of_role|>assistant<|end_of_role|>"
  },
  {
   "repo": "ibm-granite/granite-3.3-8b-instruct",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/ibm-granite/granite-3.3-8b-instruct/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "0bea7711316005a1",
   "shape": "C_0131_no_system",
   "args": "dict",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "\"function\",\n        \"function\": {\n            \"name\": \"recall_context\",\n            \"description\": \"Recall saved memories.\",\n            \"parameters\": {\n                \"type\": \"object\",\n                \"properties\": {\n                    \"query\": {\n                        \"type\": \"string\"\n                    }\n                },\n                \"required\": []\n            }\n        }\n    }\n]<|end_of_text|>\n<|start_of_role|>user<|end_of_role|>MV2USER hello, what do you remember about me?<|end_of_text|>\n<|start_of_role|>assistant<|end_of_role|><|end_of_text|>\n<|start_of_role|>tool<|end_of_role|>MV2CTX memory: the user likes short answers<|end_of_text|>\n<|start_of_role|>assistant<|end_of_role|>"
  },
  {
   "repo": "ibm-granite/granite-3.3-8b-instruct",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/ibm-granite/granite-3.3-8b-instruct/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "0bea7711316005a1",
   "shape": "D_0131_content_null",
   "args": "dict",
   "id": "call_hex24",
   "ok": false,
   "error_type": "TypeError",
   "error": "can only concatenate str (not \"NoneType\") to str"
  },
  {
   "repo": "ibm-granite/granite-3.3-8b-instruct",
   "fetched": true,
   "file": "tokenizer_config.json",
   "url": "https://huggingface.co/ibm-granite/granite-3.3-8b-instruct/resolve/main/tokenizer_config.json",
   "template": "default",
   "sha16": "0bea7711316005a1",
   "shape": "E_0131_empty_user",
   "args": "dict",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": false,
   "ctx_in_prompt": true,
   "tail": "_tools<|end_of_role|>[\n    {\n        \"type\": \"function\",\n        \"function\": {\n            \"name\": \"recall_context\",\n            \"description\": \"Recall saved memories.\",\n            \"parameters\": {\n                \"type\": \"object\",\n                \"properties\": {\n                    \"query\": {\n                        \"type\": \"string\"\n                    }\n                },\n                \"required\": []\n            }\n        }\n    }\n]<|end_of_text|>\n<|start_of_role|>user<|end_of_role|><|end_of_text|>\n<|start_of_role|>assistant<|end_of_role|><|end_of_text|>\n<|start_of_role|>tool<|end_of_role|>MV2CTX memory: the user likes short answers<|end_of_text|>\n<|start_of_role|>assistant<|end_of_role|>"
  },
  {
   "repo": "moonshotai/Kimi-K2-Instruct",
   "fetched": true,
   "file": "chat_template.jinja",
   "url": "https://huggingface.co/moonshotai/Kimi-K2-Instruct/resolve/main/chat_template.jinja",
   "template": "default",
   "sha16": "39e8c195b474c1a4",
   "shape": "A_0130_system_prompt",
   "args": "dict",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "<|im_system|>tool_declare<|im_middle|>[{\"type\":\"function\",\"function\":{\"name\":\"recall_context\",\"description\":\"Recall saved memories.\",\"parameters\":{\"type\":\"object\",\"properties\":{\"query\":{\"type\":\"string\"}},\"required\":[]}}}]<|im_end|><|im_system|>system<|im_middle|>You are a helpful assistant.\n\nMV2CTX memory: the user likes short answers<|im_end|><|im_user|>user<|im_middle|>MV2USER hello, what do you remember about me?<|im_end|><|im_assistant|>assistant<|im_middle|>"
  },
  {
   "repo": "moonshotai/Kimi-K2-Instruct",
   "fetched": true,
   "file": "chat_template.jinja",
   "url": "https://huggingface.co/moonshotai/Kimi-K2-Instruct/resolve/main/chat_template.jinja",
   "template": "default",
   "sha16": "39e8c195b474c1a4",
   "shape": "B_0131_hidden_tool",
   "args": "dict",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "\"name\":\"recall_context\",\"description\":\"Recall saved memories.\",\"parameters\":{\"type\":\"object\",\"properties\":{\"query\":{\"type\":\"string\"}},\"required\":[]}}}]<|im_end|><|im_system|>system<|im_middle|>You are a helpful assistant.<|im_end|><|im_user|>user<|im_middle|>MV2USER hello, what do you remember about me?<|im_end|><|im_assistant|>assistant<|im_middle|><|tool_calls_section_begin|><|tool_call_begin|>call_0123456789abcdef01234567<|tool_call_argument_begin|>{\"query\": \"user\"}<|tool_call_end|><|tool_calls_section_end|><|im_end|><|im_system|>recall_context<|im_middle|>## Return of call_0123456789abcdef01234567\nMV2CTX memory: the user likes short answers<|im_end|><|im_assistant|>assistant<|im_middle|>"
  },
  {
   "repo": "moonshotai/Kimi-K2-Instruct",
   "fetched": true,
   "file": "chat_template.jinja",
   "url": "https://huggingface.co/moonshotai/Kimi-K2-Instruct/resolve/main/chat_template.jinja",
   "template": "default",
   "sha16": "39e8c195b474c1a4",
   "shape": "B_0131_hidden_tool",
   "args": "dict",
   "id": "uuid4",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "_context\",\"description\":\"Recall saved memories.\",\"parameters\":{\"type\":\"object\",\"properties\":{\"query\":{\"type\":\"string\"}},\"required\":[]}}}]<|im_end|><|im_system|>system<|im_middle|>You are a helpful assistant.<|im_end|><|im_user|>user<|im_middle|>MV2USER hello, what do you remember about me?<|im_end|><|im_assistant|>assistant<|im_middle|><|tool_calls_section_begin|><|tool_call_begin|>3f2b8c1e-9a47-4d2e-b6f1-0c5d7e8a9b10<|tool_call_argument_begin|>{\"query\": \"user\"}<|tool_call_end|><|tool_calls_section_end|><|im_end|><|im_system|>recall_context<|im_middle|>## Return of 3f2b8c1e-9a47-4d2e-b6f1-0c5d7e8a9b10\nMV2CTX memory: the user likes short answers<|im_end|><|im_assistant|>assistant<|im_middle|>"
  },
  {
   "repo": "moonshotai/Kimi-K2-Instruct",
   "fetched": true,
   "file": "chat_template.jinja",
   "url": "https://huggingface.co/moonshotai/Kimi-K2-Instruct/resolve/main/chat_template.jinja",
   "template": "default",
   "sha16": "39e8c195b474c1a4",
   "shape": "B_0131_hidden_tool",
   "args": "dict",
   "id": "alnum9",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "middle|>[{\"type\":\"function\",\"function\":{\"name\":\"recall_context\",\"description\":\"Recall saved memories.\",\"parameters\":{\"type\":\"object\",\"properties\":{\"query\":{\"type\":\"string\"}},\"required\":[]}}}]<|im_end|><|im_system|>system<|im_middle|>You are a helpful assistant.<|im_end|><|im_user|>user<|im_middle|>MV2USER hello, what do you remember about me?<|im_end|><|im_assistant|>assistant<|im_middle|><|tool_calls_section_begin|><|tool_call_begin|>a1B2c3D4e<|tool_call_argument_begin|>{\"query\": \"user\"}<|tool_call_end|><|tool_calls_section_end|><|im_end|><|im_system|>recall_context<|im_middle|>## Return of a1B2c3D4e\nMV2CTX memory: the user likes short answers<|im_end|><|im_assistant|>assistant<|im_middle|>"
  },
  {
   "repo": "moonshotai/Kimi-K2-Instruct",
   "fetched": true,
   "file": "chat_template.jinja",
   "url": "https://huggingface.co/moonshotai/Kimi-K2-Instruct/resolve/main/chat_template.jinja",
   "template": "default",
   "sha16": "39e8c195b474c1a4",
   "shape": "B_0131_hidden_tool",
   "args": "str",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "\"name\":\"recall_context\",\"description\":\"Recall saved memories.\",\"parameters\":{\"type\":\"object\",\"properties\":{\"query\":{\"type\":\"string\"}},\"required\":[]}}}]<|im_end|><|im_system|>system<|im_middle|>You are a helpful assistant.<|im_end|><|im_user|>user<|im_middle|>MV2USER hello, what do you remember about me?<|im_end|><|im_assistant|>assistant<|im_middle|><|tool_calls_section_begin|><|tool_call_begin|>call_0123456789abcdef01234567<|tool_call_argument_begin|>{\"query\": \"user\"}<|tool_call_end|><|tool_calls_section_end|><|im_end|><|im_system|>recall_context<|im_middle|>## Return of call_0123456789abcdef01234567\nMV2CTX memory: the user likes short answers<|im_end|><|im_assistant|>assistant<|im_middle|>"
  },
  {
   "repo": "moonshotai/Kimi-K2-Instruct",
   "fetched": true,
   "file": "chat_template.jinja",
   "url": "https://huggingface.co/moonshotai/Kimi-K2-Instruct/resolve/main/chat_template.jinja",
   "template": "default",
   "sha16": "39e8c195b474c1a4",
   "shape": "B_0131_hidden_tool",
   "args": "str",
   "id": "uuid4",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "_context\",\"description\":\"Recall saved memories.\",\"parameters\":{\"type\":\"object\",\"properties\":{\"query\":{\"type\":\"string\"}},\"required\":[]}}}]<|im_end|><|im_system|>system<|im_middle|>You are a helpful assistant.<|im_end|><|im_user|>user<|im_middle|>MV2USER hello, what do you remember about me?<|im_end|><|im_assistant|>assistant<|im_middle|><|tool_calls_section_begin|><|tool_call_begin|>3f2b8c1e-9a47-4d2e-b6f1-0c5d7e8a9b10<|tool_call_argument_begin|>{\"query\": \"user\"}<|tool_call_end|><|tool_calls_section_end|><|im_end|><|im_system|>recall_context<|im_middle|>## Return of 3f2b8c1e-9a47-4d2e-b6f1-0c5d7e8a9b10\nMV2CTX memory: the user likes short answers<|im_end|><|im_assistant|>assistant<|im_middle|>"
  },
  {
   "repo": "moonshotai/Kimi-K2-Instruct",
   "fetched": true,
   "file": "chat_template.jinja",
   "url": "https://huggingface.co/moonshotai/Kimi-K2-Instruct/resolve/main/chat_template.jinja",
   "template": "default",
   "sha16": "39e8c195b474c1a4",
   "shape": "B_0131_hidden_tool",
   "args": "str",
   "id": "alnum9",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "middle|>[{\"type\":\"function\",\"function\":{\"name\":\"recall_context\",\"description\":\"Recall saved memories.\",\"parameters\":{\"type\":\"object\",\"properties\":{\"query\":{\"type\":\"string\"}},\"required\":[]}}}]<|im_end|><|im_system|>system<|im_middle|>You are a helpful assistant.<|im_end|><|im_user|>user<|im_middle|>MV2USER hello, what do you remember about me?<|im_end|><|im_assistant|>assistant<|im_middle|><|tool_calls_section_begin|><|tool_call_begin|>a1B2c3D4e<|tool_call_argument_begin|>{\"query\": \"user\"}<|tool_call_end|><|tool_calls_section_end|><|im_end|><|im_system|>recall_context<|im_middle|>## Return of a1B2c3D4e\nMV2CTX memory: the user likes short answers<|im_end|><|im_assistant|>assistant<|im_middle|>"
  },
  {
   "repo": "moonshotai/Kimi-K2-Instruct",
   "fetched": true,
   "file": "chat_template.jinja",
   "url": "https://huggingface.co/moonshotai/Kimi-K2-Instruct/resolve/main/chat_template.jinja",
   "template": "default",
   "sha16": "39e8c195b474c1a4",
   "shape": "C_0131_no_system",
   "args": "dict",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "escription\":\"Recall saved memories.\",\"parameters\":{\"type\":\"object\",\"properties\":{\"query\":{\"type\":\"string\"}},\"required\":[]}}}]<|im_end|><|im_system|>system<|im_middle|>You are Kimi, an AI assistant created by Moonshot AI.<|im_end|>\n<|im_user|>user<|im_middle|>MV2USER hello, what do you remember about me?<|im_end|><|im_assistant|>assistant<|im_middle|><|tool_calls_section_begin|><|tool_call_begin|>call_0123456789abcdef01234567<|tool_call_argument_begin|>{\"query\": \"user\"}<|tool_call_end|><|tool_calls_section_end|><|im_end|><|im_system|>recall_context<|im_middle|>## Return of call_0123456789abcdef01234567\nMV2CTX memory: the user likes short answers<|im_end|><|im_assistant|>assistant<|im_middle|>"
  },
  {
   "repo": "moonshotai/Kimi-K2-Instruct",
   "fetched": true,
   "file": "chat_template.jinja",
   "url": "https://huggingface.co/moonshotai/Kimi-K2-Instruct/resolve/main/chat_template.jinja",
   "template": "default",
   "sha16": "39e8c195b474c1a4",
   "shape": "D_0131_content_null",
   "args": "dict",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": true,
   "ctx_in_prompt": true,
   "tail": "\"name\":\"recall_context\",\"description\":\"Recall saved memories.\",\"parameters\":{\"type\":\"object\",\"properties\":{\"query\":{\"type\":\"string\"}},\"required\":[]}}}]<|im_end|><|im_system|>system<|im_middle|>You are a helpful assistant.<|im_end|><|im_user|>user<|im_middle|>MV2USER hello, what do you remember about me?<|im_end|><|im_assistant|>assistant<|im_middle|><|tool_calls_section_begin|><|tool_call_begin|>call_0123456789abcdef01234567<|tool_call_argument_begin|>{\"query\": \"user\"}<|tool_call_end|><|tool_calls_section_end|><|im_end|><|im_system|>recall_context<|im_middle|>## Return of call_0123456789abcdef01234567\nMV2CTX memory: the user likes short answers<|im_end|><|im_assistant|>assistant<|im_middle|>"
  },
  {
   "repo": "moonshotai/Kimi-K2-Instruct",
   "fetched": true,
   "file": "chat_template.jinja",
   "url": "https://huggingface.co/moonshotai/Kimi-K2-Instruct/resolve/main/chat_template.jinja",
   "template": "default",
   "sha16": "39e8c195b474c1a4",
   "shape": "E_0131_empty_user",
   "args": "dict",
   "id": "call_hex24",
   "ok": true,
   "user_in_prompt": false,
   "ctx_in_prompt": true,
   "tail": "<|im_middle|>[{\"type\":\"function\",\"function\":{\"name\":\"recall_context\",\"description\":\"Recall saved memories.\",\"parameters\":{\"type\":\"object\",\"properties\":{\"query\":{\"type\":\"string\"}},\"required\":[]}}}]<|im_end|><|im_system|>system<|im_middle|>You are a helpful assistant.<|im_end|><|im_user|>user<|im_middle|><|im_end|><|im_assistant|>assistant<|im_middle|><|tool_calls_section_begin|><|tool_call_begin|>call_0123456789abcdef01234567<|tool_call_argument_begin|>{\"query\": \"user\"}<|tool_call_end|><|tool_calls_section_end|><|im_end|><|im_system|>recall_context<|im_middle|>## Return of call_0123456789abcdef01234567\nMV2CTX memory: the user likes short answers<|im_end|><|im_assistant|>assistant<|im_middle|>"
  }
 ]
}