MaaS_Cl_4.X
基础信息
-
基础URL：https://genaiapi.cloudsway.net
-
API请求端点：
POST -
身份验证：通过HTTP Bearer Token方式验证，需要在请求头中携带API密钥。
| 参数名 | 类型 | 必填 | 描述 |
|---|---|---|---|
Content-Type |
string | 是 | 固定为 application/json |
Authorization |
string | 是 | Bearer {your_api_key} |
核心参数说明
| 参数 | 类型 | 必选 | 描述 |
|---|---|---|---|
model |
string | 是 | 要使用的模型ID |
messages |
array | 是 | 对话消息列表，包含role（user或assistant）和content |
max_tokens |
integer | 是 | 生成的最大token数 |
system |
string | 否 | 系统提示词，用于设定助手的行为和背景 |
temperature |
number | 否 | 采样温度，范围0-1。较高的值使输出更随机，较低的值使输出更确定 |
top_p |
number | 否 | 核采样参数，模型考虑具有top_p概率质量的token的结果 |
top_k |
integer | 否 | 仅从每个步骤中概率最高的K个token中采样 |
stream |
boolean | 否 | 是否启用流式传输，默认为false |
stop_sequences |
array | 否 | 自定义停止序列，当模型生成这些序列时停止 |
请求与响应示例
1. 原生请求调用方式
/claude/chat/completions
thinking模式
curl --location 'https://genaiapi.cloudsway.net/v2/ai/{YOUR_ENDPOINT}/claude/chat/completions' \
--header 'Authorization: Bearer {your_api_key}' \
--header 'Content-Type: application/json' \
--data-raw '{
"model": "claude-sonnet-4@20250514",
"max_tokens": 2048,
"stream":false,
"messages": [
{
"role": "user",
"content": "证明勾股定理"
}
],
"thinking": { "type": "enabled", "budget_tokens": 2000 }
}'
/v1/message
Curl 请求方式
curl --request POST \
--url https://genaiapi.cloudsway.net/{YOUR_ENDPOINT}/v1/messages \
--header 'Authorization: Bearer {your_api_key}' \
--header 'Content-Type: application/json' \
--data '{
"max_tokens": 5000,
"stream": false,
"system": [
{
"type": "text",
"text": "You are an AI assistant tasked with analyzing literary works"
}
],
"messages": [
{
"role": "user",
"content": "Analyze the major themes in Pride and Prejudice."
}
]
}'
Python 请求示例
import anthropic
# 初始化客户端（使用你提供的配置）
client = anthropic.Anthropic(
base_url="https://genaiapi.cloudsway.net/{YOUR_ENDPOINT}",
api_key="{your_api_key}"
)
# 发送消息
message = client.messages.create(
model="claude-3-5-sonnet-20241022",
max_tokens=1024,
temperature=0.7,
system="你是一个专业的编程导师，回答问题简洁明了。",
messages=[
{
"role": "user",
"content": "解释一下async/await和Promise的区别"
}
]
)
# 打印响应
print(message.content[0].text)
响应示例
{
"id": "msg_01XyZabcDeFgHiJkLmNoPqRs",
"type": "message",
"role": "assistant",
"model": "claude-3-5-sonnet-20241022",
"content": [
{
"type": "text",
"text": "async/await 是 Promise 的语法糖，主要区别在于：\n\n1. **语法简洁性**：async/await 让异步代码看起来更像同步代码，避免了 Promise 的链式调用。\n\n2. **错误处理**：async/await 可以使用 try/catch 捕获异常，而 Promise 使用 .catch()。\n\n3. **调试体验**：async/await 的调试体验更好，可以像同步代码一样设置断点。\n\n4. **本质上**：async 函数总是返回一个 Promise，await 等待 Promise 解析。\n\n示例对比：\n```javascript\n// Promise\nfetchData()\n .then(data => process(data))\n .then(result => console.log(result))\n .catch(err => console.error(err));\n\n// async/await\ntry {\n const data = await fetchData();\n const result = await process(data);\n console.log(result);\n} catch (err) {\n console.error(err);\n}\n```"
}
],
"stop_reason": "end_turn",
"stop_sequence": null,
"usage": {
"input_tokens": 25,
"output_tokens": 189
}
}
2. 兼容Openai调用方式
默认的max_tokens/max_completion_tokens是4096
开启thinking模式：
curl --location 'https://genaiapi.cloudsway.net/v1/ai/{YOUR_ENDPOINT}/chat/completions' \
--header 'Authorization: Bearer {your_api_key}' \
--header 'Content-Type: application/json' \
--data-raw '{
"model": "claude-sonnet-4@20250514",
"max_tokens": 2048,
"stream":false,
"messages": [
{
"role": "user",
"content": "证明勾股定理"
}
],
"thinking": { "type": "enabled", "budget_tokens": 2000 }
}'
关闭thinking模式
curl --location 'https://genaiapi.cloudsway.net/v1/ai/{YOUR_ENDPOINT}/chat/completions' \
--header 'Authorization: Bearer {your_api_key}' \
--header 'Content-Type: application/json' \
--data-raw '{
"model": "claude-sonnet-4@20250514",
"max_tokens": 2048,
"stream":false,
"messages": [
{
"role": "user",
"content": "证明勾股定理"
}
],
"thinking": { "type": "disabled"}
}'
3. 多轮对话
Curl 请求示例
curl -X POST "https://genaiapi.cloudsway.net/{YOUR_ENDPOINT}/v1/messages/" \
-H "Content-Type: application/json" \
-H "x-api-key: {your_api_key}" \
-H "anthropic-version: 2023-06-01" \
-d '{
"model": "claude-3-5-sonnet-20241022",
"max_tokens": 1024,
"messages": [
{
"role": "user",
"content": "Python中如何读取文件？"
},
{
"role": "assistant",
"content": "使用 `open()` 函数配合 `with` 语句：\n```python\nwith open('\''file.txt'\'', '\''r'\'') as f:\n content = f.read()\n```"
},
{
"role": "user",
"content": "那如何写入文件呢？"
}
]
}'
Python 请求示例
import anthropic
# 初始化客户端
client = anthropic.Anthropic(
base_url="https://genaiapi.cloudsway.net/{YOUR_ENDPOINT}",
api_key="{your_api_key}"
)
# 发送包含对话历史的消息
message = client.messages.create(
model="claude-3-5-sonnet-20241022",
max_tokens=1024,
messages=[
{
"role": "user",
"content": "Python中如何读取文件？"
},
{
"role": "assistant",
"content": "使用 `open()` 函数配合 `with` 语句：\n```python\nwith open('file.txt', 'r') as f:\n content = f.read()\n```"
},
{
"role": "user",
"content": "那如何写入文件呢？"
}
]
)
# 打印响应
print(message.content[0].text)
响应示例
{
"id": "msg_02AbCdEfGhIjKlMnOpQrStUv",
"type": "message",
"role": "assistant",
"model": "claude-3-5-sonnet-20241022",
"content": [
{
"type": "text",
"text": "写入文件同样使用 `open()` 函数，但模式不同：\n\n```python\n# 写入模式 ('w') - 覆盖原有内容\nwith open('file.txt', 'w') as f:\n f.write('Hello, World!')\n\n# 追加模式 ('a') - 在末尾添加内容\nwith open('file.txt', 'a') as f:\n f.write('\\nNew line')\n\n# 读写模式 ('r+') - 同时读写\nwith open('file.txt', 'r+') as f:\n content = f.read()\n f.write('\\nAppended')\n```\n\n注意：写入模式会创建文件（如果不存在），但会覆盖已存在的文件。"
}
],
"stop_reason": "end_turn",
"stop_sequence": null,
"usage": {
"input_tokens": 78,
"output_tokens": 145
}
}
4. function call (tool call)
第一次请求
curl 'https://genaiapi.cloudsway.net/{YOUR_ENDPOINT}/v1/messages' \
--header 'Authorization: Bearer {YOUR_AK}' \
--header 'Content-Type: application/json' \
--data '{
"model": "claude-opus-4-20250514",
"max_tokens": 4000,
"tools": [
{
"name": "get_weather",
"description": "Get the current weather in a given location",
"input_schema": {
"type": "object",
"properties": {
"location": {
"type": "string",
"description": "The city and state, e.g. San Francisco, CA"
},
"unit": {
"type": "string",
"enum": [
"celsius",
"fahrenheit"
],
"description": "The unit of temperature, either \"celsius\" or \"fahrenheit\""
}
},
"required": [
"location"
]
}
}
],
"messages": [
{
"role": "user",
"content": "What is the weather like in San Francisco?"
}
],
"thinking": {
"type": "enabled",
"budget_tokens": 2000
}
}'
响应内容
{
"role": "assistant",
"usage": {
"input_tokens": 471,
"cache_read_input_tokens": 0,
"output_tokens": 196,
"cache_creation_input_tokens": 0
},
"stop_reason": "tool_use",
"type": "message",3
"content": [
{
"signature": "ErcBCkgIBRACGAIiQLGZAuTsAGoReXb5s//lmeAU8yxEAzyzbNCDVVL+bkCD5sefp82OhZmg8LaQrNMDDIOZwcUTKXwWGZ8W+cNZOsESDJK/giGxwNOIo2rIWxoMVsBltOlb7IYyfzNHIjAhd0Vi5lTbgppuxotqIM1nWO8kxbAwmTkPmv2C7BkM78PBC1cnNvhXTrXmmYiLTJEqHVDoJWO3Jcd8mdw8pVHIlz4k4qkJCuLv6OOUWyAbGAI=",
"type": "thinking",
"thinking": "The user is asking about the weather in San Francisco. I can use the `get_weather` function to provide this information.\n\nLet's check the required parameters for `get_weather`:\n1. `location` (required): The user mentioned \"San Francisco\", so I'll use \"San Francisco, CA\" as the complete location.\n2. `unit` (optional): The user didn't specify a temperature unit preference, so I'll use the default unit provided by the API.\n\nSince I have all the required parameters, I can proceed with the function call."
},
{
"type": "text",
"text": "I can check the current weather in San Francisco for you."
},
{
"type": "tool_use",
"input": {
"location": "San Francisco, CA"
},
"name": "get_weather",
"id": "toolu_vrtx_01UYzvMjCpksvN4NN5jCgsMa"
}
],
"model": "claude-3-7-sonnet-20250219",
"id": "msg_vrtx_01PitSgaZ8apRctfAD8gvLYM"
}
替换第二次请求后
{
"signature": "ErcBCkgIBRACGAIiQLGZAuTsAGoReXb5s//lmeAU8yxEAzyzbNCDVVL+bkCD5sefp82OhZmg8LaQrNMDDIOZwcUTKXwWGZ8W+cNZOsESDJK/giGxwNOIo2rIWxoMVsBltOlb7IYyfzNHIjAhd0Vi5lTbgppuxotqIM1nWO8kxbAwmTkPmv2C7BkM78PBC1cnNvhXTrXmmYiLTJEqHVDoJWO3Jcd8mdw8pVHIlz4k4qkJCuLv6OOUWyAbGAI=",
"type": "thinking",
"thinking": "The user is asking about the weather in San Francisco. I can use the `get_weather` function to provide this information.\n\nLet's check the required parameters for `get_weather`:\n1. `location` (required): The user mentioned \"San Francisco\", so I'll use \"San Francisco, CA\" as the complete location.\n2. `unit` (optional): The user didn't specify a temperature unit preference, so I'll use the default unit provided by the API.\n\nSince I have all the required parameters, I can proceed with the function call."
},
替换id，tool_use_id
```plain text "id": "toolu_vrtx_01UYzvMjCpksvN4NN5jCgsMa"
```sql
curl 'https://genaiapi.cloudsway.net/{endpoint}/v1/messages' \
--header 'Authorization: Bearer {your_api_key}' \
--header 'Content-Type: application/json' \
--data '{
"model": "claude-opus-4-20250514",
"max_tokens": 4000,
"thinking": {
"type": "enabled",
"budget_tokens": 2000
},
"tools": [
{
"name": "get_weather",
"description": "Get the current weather in a given location",
"input_schema": {
"type": "object",
"properties": {
"location": {
"type": "string",
"description": "The city and state, e.g. San Francisco, CA"
},
"unit": {
"type": "string",
"enum": [
"celsius",
"fahrenheit"
],
"description": "The unit of temperature, either \"celsius\" or \"fahrenheit\""
}
},
"required": [
"location"
]
}
}
],
"messages": [
{
"role": "user",
"content": "What is the weather like in San Francisco?"
},
{
"role": "assistant",
"content": [
{
"signature": "ErcBCkgIBRACGAIiQLGZAuTsAGoReXb5s//lmeAU8yxEAzyzbNCDVVL+bkCD5sefp82OhZmg8LaQrNMDDIOZwcUTKXwWGZ8W+cNZOsESDJK/giGxwNOIo2rIWxoMVsBltOlb7IYyfzNHIjAhd0Vi5lTbgppuxotqIM1nWO8kxbAwmTkPmv2C7BkM78PBC1cnNvhXTrXmmYiLTJEqHVDoJWO3Jcd8mdw8pVHIlz4k4qkJCuLv6OOUWyAbGAI=",
"type": "thinking",
"thinking": "The user is asking about the weather in San Francisco. I can use the `get_weather` function to provide this information.\n\nLet'\''s check the required parameters for `get_weather`:\n1. `location` (required): The user mentioned \"San Francisco\", so I'\''ll use \"San Francisco, CA\" as the complete location.\n2. `unit` (optional): The user didn'\''t specify a temperature unit preference, so I'\''ll use the default unit provided by the API.\n\nSince I have all the required parameters, I can proceed with the function call."
},
{
"type": "text",
"text": "I can check the current weather in San Francisco for you."
},
{
"type": "tool_use",
"input": {
"location": "San Francisco, CA"
},
"name": "get_weather",
"id": "toolu_vrtx_01UYzvMjCpksvN4NN5jCgsMa"
}
]
},
{
"role": "user",
"content": [
{
"type": "tool_result",
"tool_use_id": "toolu_vrtx_01UYzvMjCpksvN4NN5jCgsMa",
"content": "19 degrees"
}
]
}
]
}'
5. 视觉识别（图像输入）
请求示例
Curl 请求
curl -X POST "https://genaiapi.cloudsway.net/{endpoint}/v1/messages/" \
-H "Content-Type: application/json" \
-H "x-api-key: {your_api_key}" \
-H "anthropic-version: 2023-06-01" \
-d '{
"model": null,
"max_tokens": 1024,
"messages": [
{
"role": "user",
"content": [
{
"type": "image",
"source": {
"type": "base64",
"media_type": "image/jpeg",
"data": "/9j/4AAQSkZJRgABAQAAAQABAAD/2wBDAAg..."
}
},
{
"type": "text",
"text": "这张图片里有什么？请详细描述"
}
]
}
]
}'
Python 请求
import anthropic
import base64
# 初始化客户端
client = anthropic.Anthropic(
base_url="https://genaiapi.cloudsway.net/{endpoint}",
api_key="{your_api_key}"
)
# 方式1：直接使用base64编码的图像数据
image_base64 = "/9j/4AAQSkZJRgABAQAAAQABAAD/2wBDAAg..." # 你的base64图像数据
message = client.messages.create(
model="claude-3-5-sonnet-20241022",
max_tokens=1024,
messages=[
{
"role": "user",
"content": [
{
"type": "image",
"source": {
"type": "base64",
"media_type": "image/jpeg",
"data": image_base64
}
},
{
"type": "text",
"text": "这张图片里有什么？请详细描述"
}
]
}
]
)
print(message.content[0].text)
响应示例
{
"id": "msg_05StUvWxYzAbCdEfGhIjKlMn",
"type": "message",
"role": "assistant",
"model": ,
"content": [
{
"type": "text",
"text": "这张图片显示的是：\n\n1. **主体**：一只橙色的猫坐在木质地板上\n2. **环境**：客厅的一角，有沙发和窗户\n3. **光线**：自然光从窗户射入，形成柔和的光影\n4. **细节**：猫的眼睛是绿色的，正在看着镜头\n5. **背景**：窗外可以看到一些绿色植物\n\n这是一张清晰的家庭宠物照片，氛围温馨自然。"
}
],
"stop_reason": "end_turn",
"stop_sequence": null,
"usage": {
"input_tokens": 1567,
"output_tokens": 95
}
}
6. 原生请求interleaved_thinking 参数支持
用该功能后，模型会交错展示推理过程与计算步骤，而非先隐藏思考、直接给答案
-
该参数的核心作用是控制 “推理过程是否自然穿插在回答中”，适合需要模型 “边思考边输出” 的场景（如解题、分析类任务）。
-
若未指定该参数，Claude 可能默认采用更简洁的输出模式（直接给结论 + 关键步骤，而非完整交错的思考过程）。
curl --request POST \
--url https://genaiapi.cloudsway.net/v2/ai/{endpoint}/claude/chat/completions \
--header 'Accept: */*' \
--header 'Accept-Encoding: gzip, deflate, br' \
--header 'Authorization: Bearer {your_api_key}' \
--header 'Connection: keep-alive' \
--header 'User-Agent: PostmanRuntime-ApipostRuntime/1.1.0' \
--header 'content-type: application/json' \
--data '{
"model": "claude-sonnet-4@20250514",
"max_tokens": 2048,
"stream":false,
"features": {
"interleaved_thinking": "2025-05-14"
},
"messages": [
{
"role": "user",
"content": "证明勾股定理"
}
],
"thinking": { "type": "enabled", "budget_tokens": 2000 }
}'
返回结果
{"role":"assistant","usage":{"input_tokens":43,"cache_read_input_tokens":0,"output_tokens":1128,"cache_creation_input_tokens":0},"stop_reason":"end_turn","type":"message","content":[{"signature":"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","type":"thinking","thinking":"用户要求证明勾股定理。勾股定理是一个非常经典的几何定理，有很多种证明方法。我可以提供几种不同的证明方法，包括几何证明和代数证明。让我选择几个最经典和直观的证明方法。\n\n勾股定理：在直角三角形中，直角边的平方和等于斜边的平方。即：a² + b² = c²，其中a、b是直角边，c是斜边。"},{"type":"text","text":"我来为您提供几种经典的勾股定理证明方法：\n\n## 勾股定理表述\n**在直角三角形中，两直角边的平方和等于斜边的平方**\n即：a² + b² = c²（其中a、b为直角边，c为斜边）\n\n## 证明方法一：面积法（最直观）\n\n构造一个边长为(a+b)的正方形：\n\n```\n┌─────────────────┐\n│ │ │\n│ c │ c │ b\n│ ╱ │ ╲ │\n├────╱───┼───╲────┤\n│ ╱ a │ a ╲ │\n│╱ │ ╲ │ a \n├────────┼────────┤\n│ c │ c │\n│ ╱ │ ╲ │ b\n└─────────────────┘\n```\n\n**计算总面积的两种方法：**\n- 方法1：大正方形面积 = (a+b)²\n- 方法2：4个直角三角形 + 中间小正方形 = 4×(½ab) + c² = 2ab + c²\n\n**因此：**(a+b)² = 2ab + c²\n\n展开：a² + 2ab + b² = 2ab + c²\n\n**得到：a² + b² = c²** ✓\n\n## 证明方法二：相似三角形法\n\n在直角三角形ABC中（∠C = 90°），作高CD到斜边AB：\n\n```\n C\n │╲\n │ ╲\n │ ╲\n │ ╲\n a │ ╲ b\n │ ╲\n │ ╲\n D──────╲\n └───────╲\n A c B\n```\n\n**由相似三角形性质：**\n- △ACD ∽ △ABC ∽ △CBD\n\n**得到比例关系：**\n- AC²/AB = AD/AC，即 b²/c = AD/b\n- BC²/AB = BD/BC，即 a²/c = BD/a\n\n**因此：**\n- AD = b²/c\n- BD = a²/c\n\n**由于 AD + BD = AB = c：**\nb²/c + a²/c = c\n\n**整理得：**(a² + b²)/c = c\n\n**因此：a² + b² = c²** ✓\n\n## 证明方法三：代数法（坐标几何）\n\n建立坐标系，设直角三角形三个顶点为：\n- A(0, 0)\n- B(a, 0) \n- C(0, b)\n\n**计算各边长：**\n- |AB| = a\n- |AC| = b \n- |BC| = √[(a-0)² + (0-b)²] = √(a² + b²)\n\n**根据距离公式，斜边BC的长度为：**\nc = √(a² + b²)\n\n**因此：c² = a² + b²** ✓\n\n## 应用示例\n\n**例：**直角三角形两直角边分别为3和4，求斜边长。\n\n**解：**根据勾股定理\nc² = 3² + 4² = 9 + 16 = 25\nc = 5\n\n这就是著名的\"3-4-5\"直角三角形！\n\n勾股定理不仅是几何学的基础定理，在物理学、工程学等领域也有广泛应用。"}],"model":"MaaS_Sonnet_4","id":"msg_bdrk_01X9RbtVdwyWMr6gJ4wgmwrm"}
7. 兼容openAI非流包含思考的tool调用
第一次请求
{
"model": "claude-opus-4-20250514",
"max_tokens": 4000,
"stream": false,
"thinking": {
"type": "enabled",
"budget_tokens": 2000
},
"messages": [
{
"role": "user",
"content": "What is the weather like in San Francisco?"
}
],
"tools": [
{
"function": {
"name": "get_weather",
"description": "Get the current weather in a given location",
"parameters": {
"type": "object",
"properties": {
"location": {
"type": "string",
"description": "The city and state, e.g. San Francisco, CA"
},
"unit": {
"type": "string",
"enum": [
"celsius",
"fahrenheit"
],
"description": "The unit of temperature, either \"celsius\" or \"fahrenheit\""
}
},
"required": [
"location"
]
}
}
}
]
}
第一次请求返回内容
{
"id": "msg_bdrk_01FUe2hFxT5tVJbDP3Enh5vT",
"choices": [
{
"index": 0,
"logprobs": null,
"message": {
"role": "assistant",
"content": null,
"refusal": null,
"annotations": null,
"images": null,
"reasoning_content": "The user is asking about the weather in San Francisco. I have access to a get_weather function that can help me answer this question. \n\nLet me check the required parameters:\n- location: required - The user specified \"San Francisco\" which I should interpret as \"San Francisco, CA\"\n- unit: optional - The user didn't specify a unit, so I can proceed without it (it will likely default to something)\n\nI have all the required information to make this function call.",
"function_call": null,
"tool_calls": [
{
"id": "toolu_bdrk_01TZ4CVdLzN1qHvL4gKxYRen",
"type": "function",
"function": {
"arguments": "{\"location\": \"San Francisco, CA\"}",
"name": "get_weather"
},
"custom": null
}
],
"reasoning_details": [
{
"type": "thinking",
"text": "The user is asking about the weather in San Francisco. I have access to a get_weather function that can help me answer this question. \n\nLet me check the required parameters:\n- location: required - The user specified \"San Francisco\" which I should interpret as \"San Francisco, CA\"\n- unit: optional - The user didn't specify a unit, so I can proceed without it (it will likely default to something)\n\nI have all the required information to make this function call.",
"signature": "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"
}
]
},
"finish_reason": "tool_calls",
"native_finish_reason": "tool_use"
}
],
"created": 1760448766,
"model": "MaaS_Cl_sonnet_4.5_20250929",
"object": "chat.completion",
"system_fingerprint": null,
"service_tier": null,
"usage": {
"prompt_tokens": 654,
"completion_tokens": 167,
"total_tokens": 821,
"completion_tokens_details": null,
"prompt_tokens_details": {
"audio_tokens": 0,
"cached_tokens": 0
},
"cache_creation_input_tokens": 0,
"cache_creation": null,
"gemini_cache_tokens_details": null
}
}
第二次请求
{
"model": "claude-opus-4-20250514",
"max_tokens": 4000,
"stream": false,
"thinking": {
"type": "enabled",
"budget_tokens": 2000
},
"messages": [
{
"role": "user",
"content": "What is the weather like in San Francisco?"
},
{
"role": "assistant",
"content": null,
"reasoning_content": "The user is asking about the weather in San Francisco. I have access to a get_weather function that can help me answer this question. \n\nLet me check the required parameters:\n- location: required - The user specified \"San Francisco\" which I should interpret as \"San Francisco, CA\"\n- unit: optional - The user didn't specify a unit, so I can proceed without it (it will likely default to something)\n\nI have all the required information to make this function call.",
"function_call": null,
"tool_calls": [
{
"id": "toolu_bdrk_01TZ4CVdLzN1qHvL4gKxYRen",
"type": "function",
"function": {
"arguments": "{\"location\": \"San Francisco, CA\"}",
"name": "get_weather"
},
"custom": null
}
],
"reasoning_details": [
{
"type": "thinking",
"text": "The user is asking about the weather in San Francisco. I have access to a get_weather function that can help me answer this question. \n\nLet me check the required parameters:\n- location: required - The user specified \"San Francisco\" which I should interpret as \"San Francisco, CA\"\n- unit: optional - The user didn't specify a unit, so I can proceed without it (it will likely default to something)\n\nI have all the required information to make this function call.",
"signature": "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"
}
]
},
{
"role": "tool",
"content": "28 degrees",
"tool_call_id": "toolu_bdrk_01TZ4CVdLzN1qHvL4gKxYRen"
}
],
"tools": [
{
"function": {
"name": "get_weather",
"description": "Get the current weather in a given location",
"parameters": {
"type": "object",
"properties": {
"location": {
"type": "string",
"description": "The city and state, e.g. San Francisco, CA"
},
"unit": {
"type": "string",
"enum": [
"celsius",
"fahrenheit"
],
"description": "The unit of temperature, either \"celsius\" or \"fahrenheit\""
}
},
"required": [
"location"
]
}
}
}
]
}
8. SDK调用 （Thinking 不支持SDK仅支持原生）
import anthropic
client = anthropic.Anthropic(base_url="https://genaiapi.cloudsway.net/xxx",
api_key="xxx")
message = client.beta.messages.create(
model="xxx",
max_tokens=4096,
system="你是一个专家",
messages=[{"role": "user", "content": "给我一个三百字的作文"}],
stream=True,
thinking={
"type": "enabled",
"budget_tokens": 4000
}
)
for chunk in message:
print(chunk)