Upload folder using huggingface_hub

README.md

@@ -159,8 +159,8 @@ We look forward to your feedback and to collaborating with developers and resear
 
 ## How to Use
 
-- **MiniMax Agent**: Our general agent product, built on MiniMax-M2, is now publicly available and free for a limited time: https://agent.minimaxi.com/
+- **MiniMax Agent**: Our general agent product, built on MiniMax-M2, is now publicly available and free for a limited time: https://agent.minimax.io/
 
-- **MiniMax Open Platform**: https://www.minimaxi.com/
+- **MiniMax Open Platform**: https://www.minimax.io/
 
 - **MiniMax-M2 Model Weights**: The open-source model weights are available on Hugging Face: https://huggingface.co/MiniMaxAI/

docs/function_call_guide.md

@@ -0,0 +1,482 @@
+# MiniMax-M2 函数调用（Function Call）功能指南
+
+## 简介
+
+MiniMax-M2 模型支持函数调用功能，使模型能够识别何时需要调用外部函数，并以结构化格式输出函数调用参数。本文档详细介绍了如何使用 MiniMax-M2 的函数调用功能。
+
+## 基础示例
+
+以下 Python 脚本基于 OpenAI SDK 实现了一个天气查询函数的调用示例：
+
+```python
+from openai import OpenAI
+import json
+
+client = OpenAI(base_url="http://localhost:8000/v1", api_key="dummy")
+
+def get_weather(location: str, unit: str):
+    return f"Getting the weather for {location} in {unit}..."
+
+tool_functions = {"get_weather": get_weather}
+
+tools = [{
+    "type": "function",
+    "function": {
+        "name": "get_weather",
+        "description": "Get the current weather in a given location",
+        "parameters": {
+            "type": "object",
+            "properties": {
+                "location": {"type": "string", "description": "City and state, e.g., 'San Francisco, CA'"},
+                "unit": {"type": "string", "enum": ["celsius", "fahrenheit"]}
+            },
+            "required": ["location", "unit"]
+        }
+    }
+}]
+
+response = client.chat.completions.create(
+    model=client.models.list().data[0].id,
+    messages=[{"role": "user", "content": "What's the weather like in San Francisco? use celsius."}],
+    tools=tools,
+    tool_choice="auto"
+)
+
+print(response)
+
+tool_call = response.choices[0].message.tool_calls[0].function
+print(f"Function called: {tool_call.name}")
+print(f"Arguments: {tool_call.arguments}")
+print(f"Result: {get_weather(**json.loads(tool_call.arguments))}")
+```
+
+**输出示例：**
+```
+Function called: get_weather
+Arguments: {"location": "San Francisco, CA", "unit": "celsius"}
+Result: Getting the weather for San Francisco, CA in celsius...
+```
+
+## 手动解析模型输出
+
+如果您无法使用已支持 MiniMax-M2 的推理引擎的内置解析器，或者需要使用其他推理框架（如 transformers、TGI 等），可以使用以下方法手动解析模型的原始输出。这种方法需要您自己解析模型输出的 XML 标签格式。
+
+### 使用 Transformers 的示例
+
+以下是使用 transformers 库的完整示例：
+
+```python
+from transformers import AutoTokenizer
+
+def get_default_tools():
+    return [
+        {
+          "name": "get_current_weather",
+          "description": "Get the latest weather for a location",
+          "parameters": {
+              "type": "object", 
+              "properties": {
+                  "location": {
+                      "type": "string", 
+                      "description": "A certain city, such as Beijing, Shanghai"
+                  }
+              }, 
+          }
+          "required": ["location"],
+          "type": "object"
+        }
+    ]
+
+# 加载模型和分词器
+tokenizer = AutoTokenizer.from_pretrained(model_id)
+prompt = "What's the weather like in Shanghai today?"
+messages = [
+    {"role": "system", "content": "You are a helpful assistant."},
+    {"role": "user", "content": prompt},
+]
+
+# 启用函数调用工具
+tools = get_default_tools()
+
+# 应用聊天模板，并加入工具定义
+text = tokenizer.apply_chat_template(
+    messages,
+    tokenize=False,
+    add_generation_prompt=True,
+    tools=tools
+)
+
+# 发送请求（这里使用任何推理服务）
+import requests
+payload = {
+    "model": "MiniMaxAI/MiniMax-M2",
+    "prompt": text,
+    "max_tokens": 4096
+}
+response = requests.post(
+    "http://localhost:8000/v1/completions",
+    headers={"Content-Type": "application/json"},
+    json=payload,
+    stream=False,
+)
+
+# 模型输出需要手动解析
+raw_output = response.json()["choices"][0]["text"]
+print("原始输出:", raw_output)
+
+# 使用下面的解析函数处理输出
+function_calls = parse_tool_calls(raw_output, tools)
+```
+
+## 🛠️ 函数调用的定义
+
+### 函数结构体
+
+函数调用需要在请求体中定义 `tools` 字段，每个函数由以下部分组成：
+
+```json
+{
+  "tools": [
+    {
+      "name": "search_web",
+      "description": "搜索函数。",
+      "parameters": {
+        "properties": {
+          "query_list": {
+            "description": "进行搜索的关键词，列表元素个数为1。",
+            "items": { "type": "string" },
+            "type": "array"
+          },
+          "query_tag": {
+            "description": "query的分类",
+            "items": { "type": "string" },
+            "type": "array"
+          }
+        },
+        "required": [ "query_list", "query_tag" ],
+        "type": "object"
+      }
+    }
+  ]
+}
+```
+
+**字段说明：**
+- `name`: 函数名称
+- `description`: 函数功能描述
+- `parameters`: 函数参数定义
+  - `properties`: 参数属性定义，key 是参数名，value 包含参数的详细描述
+  - `required`: 必填参数列表
+  - `type`: 参数类型（通常为 "object"）
+
+### 模型内部处理格式
+
+在 MiniMax-M2 模型内部处理���，函数定义会被转换为特殊格式并拼接到输入文本中。以下是一个完整的示例：
+
+```
+]~!b[]~b]system
+You are a helpful assistant.
+
+# Tools
+You may call one or more tools to assist with the user query.
+Here are the tools available in JSONSchema format:
+
+<tools>
+<tool>{"name": "search_web", "description": "搜索函数。", "parameters": {"type": "object", "properties": {"query_list": {"type": "array", "items": {"type": "string"}, "description": "进行搜索的关键词，列表元素个数为1。"}, "query_tag": {"type": "array", "items": {"type": "string"}, "description": "query的分类"}}, "required": ["query_list", "query_tag"]}}</tool>
+</tools>
+
+When making tool calls, use XML format to invoke tools and pass parameters:
+
+<minimax:tool_call>
+<invoke name="tool-name-1">
+<parameter name="param-key-1">param-value-1</parameter>
+<parameter name="param-key-2">param-value-2</parameter>
+...
+</invoke>
+[e~[
+]~b]user
+OpenAI 和 Gemini 的最近一次发布会都是什么时候?[e~[
+]~b]ai
+<think>
+```
+
+**格式说明：**
+
+- `]~!b[]~b]system`: System 消息开始标记
+- `[e~[`: 消息结束标记
+- `]~b]user`: User 消息开始标记
+- `]~b]ai`: Assistant 消息开始标记
+- `]~b]tool`: Tool 结果消息开始标记
+- `<tools>...</tools>`: 工具定义区域，每个工具用 `<tool>` 标签包裹，内容为 JSON Schema
+- `<minimax:tool_call>...</minimax:tool_call>`: 工具调用区域
+- `<think>`: 生成时的思考过程标记（可选）
+
+### 模型输出格式
+
+MiniMax-M2使用结构化的 XML 标签格式：
+
+```xml
+<minimax:tool_call>
+<invoke name="search_web">
+<parameter name="query_tag">["technology", "events"]</parameter>
+<parameter name="query_list">["\"OpenAI\" \"latest\" \"release\""]</parameter>
+</invoke>
+<invoke name="search_web">
+<parameter name="query_tag">["technology", "events"]</parameter>
+<parameter name="query_list">["\"Gemini\" \"latest\" \"release\""]</parameter>
+</invoke>
+</minimax:tool_call>
+```
+
+每个函数调用使用 `<invoke name="函数名">` 标签，参数使用 `<parameter name="参数名">` 标签包裹。
+
+## 手动解析函数调用结果
+
+### 解析函数调用
+
+MiniMax-M2使用结构化的 XML 标签，需要不同的解析方式。核心函数如下：
+
+```python
+import re
+import json
+from typing import Any, Optional, List, Dict
+
+
+def extract_name(name_str: str) -> str:
+    """从引号包裹的字符串中提取名称"""
+    name_str = name_str.strip()
+    if name_str.startswith('"') and name_str.endswith('"'):
+        return name_str[1:-1]
+    elif name_str.startswith("'") and name_str.endswith("'"):
+        return name_str[1:-1]
+    return name_str
+
+
+def convert_param_value(value: str, param_type: str) -> Any:
+    """根据参数类型转换参数值"""
+    if value.lower() == "null":
+        return None
+        
+    param_type = param_type.lower()
+    
+    if param_type in ["string", "str", "text"]:
+        return value
+    elif param_type in ["integer", "int"]:
+        try:
+            return int(value)
+        except (ValueError, TypeError):
+            return value
+    elif param_type in ["number", "float"]:
+        try:
+            val = float(value)
+            return val if val != int(val) else int(val)
+        except (ValueError, TypeError):
+            return value
+    elif param_type in ["boolean", "bool"]:
+        return value.lower() in ["true", "1"]
+    elif param_type in ["object", "array"]:
+        try:
+            return json.loads(value)
+        except json.JSONDecodeError:
+            return value
+    else:
+        # 尝试 JSON 解析，失败则返回字符串
+        try:
+            return json.loads(value)
+        except json.JSONDecodeError:
+            return value
+
+
+def parse_tool_calls(model_output: str, tools: Optional[List[Dict]] = None) -> List[Dict]:
+    """
+    从模型输出中提取所有工具调用
+    
+    Args:
+        model_output: 模型的完整输出文本
+        tools: 工具定义列表，用于获取参数类型信息，格式可以是：
+               - [{"name": "...", "parameters": {...}}]
+               - [{"type": "function", "function": {"name": "...", "parameters": {...}}}]
+    
+    Returns:
+        解析后的工具调用列表，每个元素包含 name 和 arguments 字段
+    
+    Example:
+        >>> tools = [{
+        ...     "name": "get_weather",
+        ...     "parameters": {
+        ...         "type": "object",
+        ...         "properties": {
+        ...             "location": {"type": "string"},
+        ...             "unit": {"type": "string"}
+        ...         }
+        ...     }
+        ... }]
+        >>> output = '''<minimax:tool_call>
+        ... <invoke name="get_weather">
+        ... <parameter name="location">San Francisco</parameter>
+        ... <parameter name="unit">celsius</parameter>
+        ... </invoke>
+        ... </minimax:tool_call>'''
+        >>> result = parse_tool_calls(output, tools)
+        >>> print(result)
+        [{'name': 'get_weather', 'arguments': {'location': 'San Francisco', 'unit': 'celsius'}}]
+    """
+    # 快速检查是否包含工具调用标记
+    if "<minimax:tool_call>" not in model_output:
+        return []
+    
+    tool_calls = []
+    
+    try:
+        # 匹配所有 <minimax:tool_call> 块
+        tool_call_regex = re.compile(r"<minimax:tool_call>(.*?)</minimax:tool_call>", re.DOTALL)
+        invoke_regex = re.compile(r"<invoke name=(.*?)</invoke>", re.DOTALL)
+        parameter_regex = re.compile(r"<parameter name=(.*?)</parameter>", re.DOTALL)
+        
+        # 遍历所有 tool_call 块
+        for tool_call_match in tool_call_regex.findall(model_output):
+            # 遍历该块中的所有 invoke
+            for invoke_match in invoke_regex.findall(tool_call_match):
+                # 提取函数名
+                name_match = re.search(r'^([^>]+)', invoke_match)
+                if not name_match:
+                    continue
+                
+                function_name = extract_name(name_match.group(1))
+                
+                # 获取参数配置
+                param_config = {}
+                if tools:
+                    for tool in tools:
+                        tool_name = tool.get("name") or tool.get("function", {}).get("name")
+                        if tool_name == function_name:
+                            params = tool.get("parameters") or tool.get("function", {}).get("parameters")
+                            if isinstance(params, dict) and "properties" in params:
+                                param_config = params["properties"]
+                            break
+                
+                # 提取参数
+                param_dict = {}
+                for match in parameter_regex.findall(invoke_match):
+                    param_match = re.search(r'^([^>]+)>(.*)', match, re.DOTALL)
+                    if param_match:
+                        param_name = extract_name(param_match.group(1))
+                        param_value = param_match.group(2).strip()
+                        
+                        # 去除首尾的换行符
+                        if param_value.startswith('\n'):
+                            param_value = param_value[1:]
+                        if param_value.endswith('\n'):
+                            param_value = param_value[:-1]
+                        
+                        # 获取参数类型并转换
+                        param_type = "string"
+                        if param_name in param_config:
+                            if isinstance(param_config[param_name], dict) and "type" in param_config[param_name]:
+                                param_type = param_config[param_name]["type"]
+                        
+                        param_dict[param_name] = convert_param_value(param_value, param_type)
+                
+                tool_calls.append({
+                    "name": function_name,
+                    "arguments": param_dict
+                })
+    
+    except Exception as e:
+        print(f"解析工具调用失败: {e}")
+        return []
+    
+    return tool_calls
+```
+
+**使用示例：**
+
+```python
+# 定义工具
+tools = [
+    {
+        "name": "get_weather",
+        "parameters": {
+            "type": "object",
+            "properties": {
+                "location": {"type": "string"},
+                "unit": {"type": "string"}
+            },
+            "required": ["location", "unit"]
+        }
+    }
+]
+
+# 模型输出
+model_output = """我来帮你查询天气。
+<minimax:tool_call>
+<invoke name="get_weather">
+<parameter name="location">San Francisco</parameter>
+<parameter name="unit">celsius</parameter>
+</invoke>
+</minimax:tool_call>"""
+
+# 解析工具调用
+tool_calls = parse_tool_calls(model_output, tools)
+
+# 输出结果
+for call in tool_calls:
+    print(f"调用函数: {call['name']}")
+    print(f"参数: {call['arguments']}")
+    # 输出: 调用函数: get_weather
+    #      参数: {'location': 'San Francisco', 'unit': 'celsius'}
+```
+
+### 执行函数调用
+
+解析完成后，您可以执行对应的函数并构建返回结果：
+
+```python
+def execute_function_call(function_name: str, arguments: dict):
+    """执行函数调用并返回结果"""
+    if function_name == "get_weather":
+        location = arguments.get("location", "未知位置")
+        unit = arguments.get("unit", "celsius")
+        # 构建函数执行结果
+        return {
+            "role": "tool", 
+            "content": [
+              {
+                "name": function_name,
+                "type": "text",
+                "text": json.dumps({
+                    "location": location, 
+                    "temperature": "25", 
+                    "unit": unit, 
+                    "weather": "晴朗"
+                }, ensure_ascii=False)
+              }
+            ] 
+          }
+    elif function_name == "search_web":
+        query_list = arguments.get("query_list", [])
+        query_tag = arguments.get("query_tag", [])
+        # 模拟搜索结果
+        return {
+            "role": "tool",
+            "content": [
+              {
+                "name": function_name,
+                "type": "text",
+                "text": f"搜索关键词: {query_list}, 分类: {query_tag}\n搜索结果: 相关信息已找到"
+              }
+            ]
+          }
+    
+    return None
+```
+
+### 将函数执行结果返回给模型
+
+成功解析函数调用后，您应将函数执行结果添加到对话历史中，以便模型在后续交互中能够访问和利用这些信息，拼接格式参考chat_template.jinja
+
+## 参考资料
+
+- [MiniMax-M2 模型仓库](https://github.com/MiniMaxAI/MiniMax-M2)
+- [vLLM 项目主页](https://github.com/vllm-project/vllm)
+- [OpenAI Python SDK](https://github.com/openai/openai-python)

docs/function_call_guide_en.md

@@ -0,0 +1,482 @@
+# MiniMax-M2 Function Call Guide
+
+## Introduction
+
+The MiniMax-M2 model supports function calling capabilities, enabling the model to identify when external functions need to be called and output function call parameters in a structured format. This document provides detailed instructions on how to use the function calling features of MiniMax-M2.
+
+## Basic Example
+
+The following Python script implements a weather query function call example based on the OpenAI SDK:
+
+```python
+from openai import OpenAI
+import json
+
+client = OpenAI(base_url="http://localhost:8000/v1", api_key="dummy")
+
+def get_weather(location: str, unit: str):
+    return f"Getting the weather for {location} in {unit}..."
+
+tool_functions = {"get_weather": get_weather}
+
+tools = [{
+    "type": "function",
+    "function": {
+        "name": "get_weather",
+        "description": "Get the current weather in a given location",
+        "parameters": {
+            "type": "object",
+            "properties": {
+                "location": {"type": "string", "description": "City and state, e.g., 'San Francisco, CA'"},
+                "unit": {"type": "string", "enum": ["celsius", "fahrenheit"]}
+            },
+            "required": ["location", "unit"]
+        }
+    }
+}]
+
+response = client.chat.completions.create(
+    model=client.models.list().data[0].id,
+    messages=[{"role": "user", "content": "What's the weather like in San Francisco? use celsius."}],
+    tools=tools,
+    tool_choice="auto"
+)
+
+print(response)
+
+tool_call = response.choices[0].message.tool_calls[0].function
+print(f"Function called: {tool_call.name}")
+print(f"Arguments: {tool_call.arguments}")
+print(f"Result: {get_weather(**json.loads(tool_call.arguments))}")
+```
+
+**Output Example:**
+```
+Function called: get_weather
+Arguments: {"location": "San Francisco, CA", "unit": "celsius"}
+Result: Getting the weather for San Francisco, CA in celsius...
+```
+
+## Manually Parsing Model Output
+
+If you cannot use the built-in parser of inference engines that support MiniMax-M2, or need to use other inference frameworks (such as transformers, TGI, etc.), you can manually parse the model's raw output using the following method. This approach requires you to parse the XML tag format of the model output yourself.
+
+### Example Using Transformers
+
+Here is a complete example using the transformers library:
+
+```python
+from transformers import AutoTokenizer
+
+def get_default_tools():
+    return [
+        {
+          "name": "get_current_weather",
+          "description": "Get the latest weather for a location",
+          "parameters": {
+              "type": "object", 
+              "properties": {
+                  "location": {
+                      "type": "string", 
+                      "description": "A certain city, such as Beijing, Shanghai"
+                  }
+              }, 
+          }
+          "required": ["location"],
+          "type": "object"
+        }
+    ]
+
+# Load model and tokenizer
+tokenizer = AutoTokenizer.from_pretrained(model_id)
+prompt = "What's the weather like in Shanghai today?"
+messages = [
+    {"role": "system", "content": "You are a helpful assistant."},
+    {"role": "user", "content": prompt},
+]
+
+# Enable function calling tools
+tools = get_default_tools()
+
+# Apply chat template and include tool definitions
+text = tokenizer.apply_chat_template(
+    messages,
+    tokenize=False,
+    add_generation_prompt=True,
+    tools=tools
+)
+
+# Send request (using any inference service)
+import requests
+payload = {
+    "model": "MiniMaxAI/MiniMax-M2",
+    "prompt": text,
+    "max_tokens": 4096
+}
+response = requests.post(
+    "http://localhost:8000/v1/completions",
+    headers={"Content-Type": "application/json"},
+    json=payload,
+    stream=False,
+)
+
+# Model output needs manual parsing
+raw_output = response.json()["choices"][0]["text"]
+print("Raw output:", raw_output)
+
+# Use the parsing function below to process the output
+function_calls = parse_tool_calls(raw_output, tools)
+```
+
+## 🛠️ Function Call Definition
+
+### Function Structure
+
+Function calls need to define the `tools` field in the request body. Each function consists of the following parts:
+
+```json
+{
+  "tools": [
+    {
+      "name": "search_web",
+      "description": "Search function.",
+      "parameters": {
+        "properties": {
+          "query_list": {
+            "description": "Keywords for search, list should contain 1 element.",
+            "items": { "type": "string" },
+            "type": "array"
+          },
+          "query_tag": {
+            "description": "Category of query",
+            "items": { "type": "string" },
+            "type": "array"
+          }
+        },
+        "required": [ "query_list", "query_tag" ],
+        "type": "object"
+      }
+    }
+  ]
+}
+```
+
+**Field Descriptions:**
+- `name`: Function name
+- `description`: Function description
+- `parameters`: Function parameter definition
+  - `properties`: Parameter property definition, where key is the parameter name and value contains detailed parameter description
+  - `required`: List of required parameters
+  - `type`: Parameter type (usually "object")
+
+### Internal Processing Format
+
+When processing within the MiniMax-M2 model, function definitions are converted to a special format and concatenated to the input text. Here is a complete example:
+
+```
+]~!b[]~b]system
+You are a helpful assistant.
+
+# Tools
+You may call one or more tools to assist with the user query.
+Here are the tools available in JSONSchema format:
+
+<tools>
+<tool>{"name": "search_web", "description": "Search function.", "parameters": {"type": "object", "properties": {"query_list": {"type": "array", "items": {"type": "string"}, "description": "Keywords for search, list should contain 1 element."}, "query_tag": {"type": "array", "items": {"type": "string"}, "description": "Category of query"}}, "required": ["query_list", "query_tag"]}}</tool>
+</tools>
+
+When making tool calls, use XML format to invoke tools and pass parameters:
+
+<minimax:tool_call>
+<invoke name="tool-name-1">
+<parameter name="param-key-1">param-value-1</parameter>
+<parameter name="param-key-2">param-value-2</parameter>
+...
+</invoke>
+[e~[
+]~b]user
+When were the latest announcements from OpenAI and Gemini?[e~[
+]~b]ai
+<think>
+```
+
+**Format Description:**
+
+- `]~!b[]~b]system`: System message start marker
+- `[e~[`: Message end marker
+- `]~b]user`: User message start marker
+- `]~b]ai`: Assistant message start marker
+- `]~b]tool`: Tool result message start marker
+- `<tools>...</tools>`: Tool definition area, each tool is wrapped with `<tool>` tag, content is JSON Schema
+- `<minimax:tool_call>...</minimax:tool_call>`: Tool call area
+- `<think>`: Thinking process marker during generation (optional)
+
+### Model Output Format
+
+MiniMax-M2 uses structured XML tag format:
+
+```xml
+<minimax:tool_call>
+<invoke name="search_web">
+<parameter name="query_tag">["technology", "events"]</parameter>
+<parameter name="query_list">["\"OpenAI\" \"latest\" \"release\""]</parameter>
+</invoke>
+<invoke name="search_web">
+<parameter name="query_tag">["technology", "events"]</parameter>
+<parameter name="query_list">["\"Gemini\" \"latest\" \"release\""]</parameter>
+</invoke>
+</minimax:tool_call>
+```
+
+Each function call uses the `<invoke name="function_name">` tag, and parameters use the `<parameter name="parameter_name">` tag wrapper.
+
+## Manually Parsing Function Call Results
+
+### Parsing Function Calls
+
+MiniMax-M2 uses structured XML tags, which require a different parsing approach. The core function is as follows:
+
+```python
+import re
+import json
+from typing import Any, Optional, List, Dict
+
+
+def extract_name(name_str: str) -> str:
+    """Extract name from quoted string"""
+    name_str = name_str.strip()
+    if name_str.startswith('"') and name_str.endswith('"'):
+        return name_str[1:-1]
+    elif name_str.startswith("'") and name_str.endswith("'"):
+        return name_str[1:-1]
+    return name_str
+
+
+def convert_param_value(value: str, param_type: str) -> Any:
+    """Convert parameter value based on parameter type"""
+    if value.lower() == "null":
+        return None
+        
+    param_type = param_type.lower()
+    
+    if param_type in ["string", "str", "text"]:
+        return value
+    elif param_type in ["integer", "int"]:
+        try:
+            return int(value)
+        except (ValueError, TypeError):
+            return value
+    elif param_type in ["number", "float"]:
+        try:
+            val = float(value)
+            return val if val != int(val) else int(val)
+        except (ValueError, TypeError):
+            return value
+    elif param_type in ["boolean", "bool"]:
+        return value.lower() in ["true", "1"]
+    elif param_type in ["object", "array"]:
+        try:
+            return json.loads(value)
+        except json.JSONDecodeError:
+            return value
+    else:
+        # Try JSON parsing, return string if failed
+        try:
+            return json.loads(value)
+        except json.JSONDecodeError:
+            return value
+
+
+def parse_tool_calls(model_output: str, tools: Optional[List[Dict]] = None) -> List[Dict]:
+    """
+    Extract all tool calls from model output
+    
+    Args:
+        model_output: Complete output text from the model
+        tools: Tool definition list for getting parameter type information, format can be:
+               - [{"name": "...", "parameters": {...}}]
+               - [{"type": "function", "function": {"name": "...", "parameters": {...}}}]
+    
+    Returns:
+        Parsed tool call list, each element contains name and arguments fields
+    
+    Example:
+        >>> tools = [{
+        ...     "name": "get_weather",
+        ...     "parameters": {
+        ...         "type": "object",
+        ...         "properties": {
+        ...             "location": {"type": "string"},
+        ...             "unit": {"type": "string"}
+        ...         }
+        ...     }
+        ... }]
+        >>> output = '''<minimax:tool_call>
+        ... <invoke name="get_weather">
+        ... <parameter name="location">San Francisco</parameter>
+        ... <parameter name="unit">celsius</parameter>
+        ... </invoke>
+        ... </minimax:tool_call>'''
+        >>> result = parse_tool_calls(output, tools)
+        >>> print(result)
+        [{'name': 'get_weather', 'arguments': {'location': 'San Francisco', 'unit': 'celsius'}}]
+    """
+    # Quick check if tool call marker is present
+    if "<minimax:tool_call>" not in model_output:
+        return []
+    
+    tool_calls = []
+    
+    try:
+        # Match all <minimax:tool_call> blocks
+        tool_call_regex = re.compile(r"<minimax:tool_call>(.*?)</minimax:tool_call>", re.DOTALL)
+        invoke_regex = re.compile(r"<invoke name=(.*?)</invoke>", re.DOTALL)
+        parameter_regex = re.compile(r"<parameter name=(.*?)</parameter>", re.DOTALL)
+        
+        # Iterate through all tool_call blocks
+        for tool_call_match in tool_call_regex.findall(model_output):
+            # Iterate through all invokes in this block
+            for invoke_match in invoke_regex.findall(tool_call_match):
+                # Extract function name
+                name_match = re.search(r'^([^>]+)', invoke_match)
+                if not name_match:
+                    continue
+                
+                function_name = extract_name(name_match.group(1))
+                
+                # Get parameter configuration
+                param_config = {}
+                if tools:
+                    for tool in tools:
+                        tool_name = tool.get("name") or tool.get("function", {}).get("name")
+                        if tool_name == function_name:
+                            params = tool.get("parameters") or tool.get("function", {}).get("parameters")
+                            if isinstance(params, dict) and "properties" in params:
+                                param_config = params["properties"]
+                            break
+                
+                # Extract parameters
+                param_dict = {}
+                for match in parameter_regex.findall(invoke_match):
+                    param_match = re.search(r'^([^>]+)>(.*)', match, re.DOTALL)
+                    if param_match:
+                        param_name = extract_name(param_match.group(1))
+                        param_value = param_match.group(2).strip()
+                        
+                        # Remove leading and trailing newlines
+                        if param_value.startswith('\n'):
+                            param_value = param_value[1:]
+                        if param_value.endswith('\n'):
+                            param_value = param_value[:-1]
+                        
+                        # Get parameter type and convert
+                        param_type = "string"
+                        if param_name in param_config:
+                            if isinstance(param_config[param_name], dict) and "type" in param_config[param_name]:
+                                param_type = param_config[param_name]["type"]
+                        
+                        param_dict[param_name] = convert_param_value(param_value, param_type)
+                
+                tool_calls.append({
+                    "name": function_name,
+                    "arguments": param_dict
+                })
+    
+    except Exception as e:
+        print(f"Failed to parse tool calls: {e}")
+        return []
+    
+    return tool_calls
+```
+
+**Usage Example:**
+
+```python
+# Define tools
+tools = [
+    {
+        "name": "get_weather",
+        "parameters": {
+            "type": "object",
+            "properties": {
+                "location": {"type": "string"},
+                "unit": {"type": "string"}
+            },
+            "required": ["location", "unit"]
+        }
+    }
+]
+
+# Model output
+model_output = """Let me help you query the weather.
+<minimax:tool_call>
+<invoke name="get_weather">
+<parameter name="location">San Francisco</parameter>
+<parameter name="unit">celsius</parameter>
+</invoke>
+</minimax:tool_call>"""
+
+# Parse tool calls
+tool_calls = parse_tool_calls(model_output, tools)
+
+# Output results
+for call in tool_calls:
+    print(f"Function called: {call['name']}")
+    print(f"Arguments: {call['arguments']}")
+    # Output: Function called: get_weather
+    #         Arguments: {'location': 'San Francisco', 'unit': 'celsius'}
+```
+
+### Executing Function Calls
+
+After parsing is complete, you can execute the corresponding function and construct the return result:
+
+```python
+def execute_function_call(function_name: str, arguments: dict):
+    """Execute function call and return result"""
+    if function_name == "get_weather":
+        location = arguments.get("location", "Unknown location")
+        unit = arguments.get("unit", "celsius")
+        # Build function execution result
+        return {
+            "role": "tool", 
+            "content": [
+              {
+                "name": function_name,
+                "type": "text",
+                "text": json.dumps({
+                    "location": location, 
+                    "temperature": "25", 
+                    "unit": unit, 
+                    "weather": "Sunny"
+                }, ensure_ascii=False)
+              }
+            ] 
+          }
+    elif function_name == "search_web":
+        query_list = arguments.get("query_list", [])
+        query_tag = arguments.get("query_tag", [])
+        # Simulate search results
+        return {
+            "role": "tool",
+            "content": [
+              {
+                "name": function_name,
+                "type": "text",
+                "text": f"Search keywords: {query_list}, Category: {query_tag}\nSearch results: Relevant information found"
+              }
+            ]
+          }
+    
+    return None
+```
+
+### Returning Function Execution Results to the Model
+
+After successfully parsing function calls, you should add the function execution results to the conversation history so that the model can access and utilize this information in subsequent interactions. Refer to chat_template.jinja for concatenation format.
+
+## References
+
+- [MiniMax-M2 Model Repository](https://github.com/MiniMaxAI/MiniMax-M2)
+- [vLLM Project Homepage](https://github.com/vllm-project/vllm)
+- [OpenAI Python SDK](https://github.com/openai/openai-python)

docs/vllm_deploy_guide.md

@@ -0,0 +1,94 @@
+# MiniMax M2 Model vLLM Deployment Guide
+
+We recommend using [vLLM](https://docs.vllm.ai/en/latest/) to deploy the [MiniMax-M2](https://huggingface.co/MiniMaxAI/MiniMax-M2) model. vLLM is a high-performance inference engine with excellent serving throughput, efficient and intelligent memory management, powerful batch request processing capabilities, and deeply optimized underlying performance. We recommend reviewing vLLM's official documentation to check hardware compatibility before deployment.
+
+## Applicable Models
+
+This guide applies to the following models. Simply modify the model name during deployment. The deployment process is illustrated using MiniMax-M2 as an example.
+
+- MiniMaxAI/MiniMax-M2
+
+## System Requirements
+
+- OS: Linux
+
+- Python: 3.9 - 3.12
+
+- GPU:
+
+  - compute capability 7.0 or higher
+
+  - Memory requirements: 220 GB for weights, 60 GB per 1M context tokens
+
+The following are recommended configurations; actual requirements should be adjusted based on your use case:
+
+- 4x 96GB GPUs: Supports context input of up to 400K tokens.
+
+- 8x 144GB GPUs: Supports context input of up to 3M tokens.
+
+## Deployment with Python
+
+It is recommended to use a virtual environment (such as venv, conda, or uv) to avoid dependency conflicts. We recommend installing vLLM in a fresh Python environment:
+
+```bash
+# Not yet released, please install nightly build
+uv pip install -U vllm \
+    --torch-backend=auto \
+    --extra-index-url https://wheels.vllm.ai/nightly
+# If released, install using uv
+uv pip install "vllm" --torch-backend=auto
+```
+
+Run the following command to start the vLLM server. vLLM will automatically download and cache the MiniMax-M2 model from Hugging Face.
+
+4-GPU deployment command:
+
+```bash
+SAFETENSORS_FAST_GPU=1 VLLM_USE_V1=0 vllm serve \
+    --model MiniMaxAI/MiniMax-M2 \
+    --trust-remote-code \
+    --enable-expert-parallel --tensor-parallel-size 4 \
+    --enable-auto-tool-choice --tool-call-parser minimax_m2 \
+    --reasoning-parser minimax_m2
+```
+
+## Testing Deployment
+
+After startup, you can test the vLLM OpenAI-compatible API with the following command:
+
+```bash
+curl http://localhost:8000/v1/chat/completions \
+    -H "Content-Type: application/json" \
+    -d '{
+        "model": "MiniMaxAI/MiniMax-M2",
+        "messages": [
+            {"role": "system", "content": [{"type": "text", "text": "You are a helpful assistant."}]},
+            {"role": "user", "content": [{"type": "text", "text": "Who won the world series in 2020?"}]}
+        ]
+    }'
+```
+
+## Common Issues
+
+### Hugging Face Network Issues
+
+If you encounter network issues, you can set up a proxy before pulling the model.
+
+```bash
+export HF_ENDPOINT=https://hf-mirror.com
+```
+
+### MiniMax-M2 model is not currently supported
+
+This vLLM version is outdated. Please upgrade to the latest version.
+
+## Getting Support
+
+If you encounter any issues while deploying the MiniMax model:
+
+- Contact our technical support team through official channels such as email at api@minimaxi.com
+
+- Submit an issue on our [GitHub](https://github.com/MiniMax-AI) repository
+
+We continuously optimize the deployment experience for our models. Feedback is welcome!
+

docs/vllm_deploy_guide_cn.md

@@ -0,0 +1,91 @@
+# MiniMax M2 模型 vLLM 部署指南
+
+我们推荐使用 [vLLM](https://docs.vllm.ai/en/latest/) 来部署 [MiniMax-M2](https://huggingface.co/MiniMaxAI/MiniMax-M2) 模型。vLLM 是一个高性能的推理引擎，其具有卓越的服务吞吐、高效智能的内存管理机制、强大的批量请求处理能力、深度优化的底层性能等特性。我们建议在部署之前查看 vLLM 的官方文档以检查硬件兼容性。
+
+## 本文档适用模型
+
+本文档适用以下模型，只需在部署时修改模型名称即可。以下以 MiniMax-M2 为例说明部署流程。
+
+- MiniMaxAI/MiniMax-M2
+
+## 环境要求
+
+- OS：Linux
+
+- Python：3.9 - 3.12
+
+- GPU：
+
+  - compute capability 7.0 or higher
+
+  - 显存需求：权重需要 220 GB，每 1M 上下文 token 需要 60 GB
+
+以下为推荐配置，实际需求请根据业务场景调整：
+
+- 96G x4 GPU：支持 40 万 token 的上下文输入。
+
+- 144G x8 GPU：支持长达 300 万 token 的上下文输入。
+
+## 使用 Python 部署
+
+建议使用虚拟环境（如 venv、conda、uv）以避免依赖冲突。建议在全新的 Python 环境中安装 vLLM：
+```bash
+# 尚未 release，请安装 nightly 构建
+uv pip install -U vllm \
+    --torch-backend=auto \
+    --extra-index-url https://wheels.vllm.ai/nightly
+# 如果 release，使用 uv 安装
+uv pip install "vllm" --torch-backend=auto
+```
+
+运行如下命令启动 vLLM 服务器，vLLM 会自动从 Huggingface 下载并缓存 MiniMax-M2 模型。
+
+4 卡部署命令：
+
+```bash
+SAFETENSORS_FAST_GPU=1 VLLM_USE_V1=0 vllm serve \
+    --model MiniMaxAI/MiniMax-M2 \
+    --trust-remote-code \
+    --enable-expert-parallel --tensor-parallel-size 4 \
+    --enable-auto-tool-choice --tool-call-parser minimax_m2 \
+    --reasoning-parser minimax_m2
+```
+
+## 测试部署
+
+启动后，可以通过如下命令测试 vLLM OpenAI 兼容接口：
+
+```bash
+curl http://localhost:8000/v1/chat/completions \
+    -H "Content-Type: application/json" \
+    -d '{
+        "model": "MiniMaxAI/MiniMax-M2",
+        "messages": [
+            {"role": "system", "content": [{"type": "text", "text": "You are a helpful assistant."}]},
+            {"role": "user", "content": [{"type": "text", "text": "Who won the world series in 2020?"}]}
+        ]
+    }'
+```
+
+## 常见问题
+
+### Huggingface 网络问题
+
+如果遇到网络问题，可以设置代理后再进行拉取。
+
+```bash
+export HF_ENDPOINT=https://hf-mirror.com
+```
+
+### MiniMax-M2 model is not currently supported
+
+该 vLLM 版本过旧，请升级到最新版本。
+
+## 获取支持
+
+如果在部署 MiniMax 模型过程中遇到任何问题：
+
+- 通过邮箱 api@minimaxi.com 等官方渠道联系我们的技术支持团队
+
+- 在我们的 [GitHub](https://github.com/MiniMax-AI) 仓库提交 Issue
+我们会持续优化模型的部署体验，欢迎反馈！