DeepSeek API 使用指南：聊天、推理与代码生成，OpenAI 经济替代方案

deepseek by vm0-ai/vm0-skills

124 周安装量

48 GitHub Stars

GitHub

安装命令

npx skills add https://github.com/vm0-ai/vm0-skills --skill deepseek

AI/机器学习代码生成 API

🇨🇳中文介绍

DeepSeek API

通过直接的 curl 调用使用 DeepSeek API，以访问强大的 AI 语言模型，用于聊天、推理和代码生成。

官方文档：https://api-docs.deepseek.com/

使用场景

在以下情况下使用此技能：

使用 DeepSeek-V3.2 模型进行聊天补全
使用推理模型进行深度推理任务
代码生成和补全（FIM - 中间填充）
作为OpenAI兼容API的经济高效替代方案

先决条件

在 DeepSeek 平台注册并创建账户
前往 API 密钥并生成新的 API 密钥
为您的账户充值（无免费额度，但定价非常实惠）

export DEEPSEEK_API_KEY="your-api-key"

定价（每 100 万 token）

广告位招租

在这里展示您的产品或服务

触达数万 AI 开发者，精准高效

联系我们

1. 基础聊天补全

发送简单的聊天消息：

写入 /tmp/deepseek_request.json：

{
  "model": "deepseek-chat",
  "messages": [
    {
      "role": "system",
      "content": "You are a helpful assistant."
    },
    {
      "role": "user",
      "content": "Hello, who are you?"
    }
  ]
}

curl -s "https://api.deepseek.com/chat/completions" -X POST -H "Content-Type: application/json" -H "Authorization: Bearer $(printenv DEEPSEEK_API_KEY)" -d @/tmp/deepseek_request.json

deepseek-chat：DeepSeek-V3.2 非思考模式（128K 上下文，8K 最大输出）
deepseek-reasoner：DeepSeek-V3.2 思考模式（128K 上下文，64K 最大输出）

2. 带温度控制的聊天

通过 temperature 调整创造力/随机性：

写入 /tmp/deepseek_request.json：

{
  "model": "deepseek-chat",
  "messages": [
    {
      "role": "user",
      "content": "Write a short poem about coding."
    }
  ],
  "temperature": 0.7,
  "max_tokens": 200
}

curl -s "https://api.deepseek.com/chat/completions" -X POST -H "Content-Type: application/json" -H "Authorization: Bearer $(printenv DEEPSEEK_API_KEY)" -d @/tmp/deepseek_request.json | jq -r '.choices[0].message.content'

temperature（0-2，默认 1）：值越高越有创意，越低越确定
top_p（0-1，默认 1）：核心采样阈值
max_tokens：要生成的最大 token 数

获取实时的逐 token 输出：

写入 /tmp/deepseek_request.json：

{
  "model": "deepseek-chat",
  "messages": [
    {
      "role": "user",
      "content": "Explain quantum computing in simple terms."
    }
  ],
  "stream": true
}

curl -s "https://api.deepseek.com/chat/completions" -X POST -H "Content-Type: application/json" -H "Authorization: Bearer $(printenv DEEPSEEK_API_KEY)" -d @/tmp/deepseek_request.json

流式返回服务器发送事件（SSE），包含增量数据块，以 data: [DONE] 结束。

4. 深度推理（思考模式）

使用推理器模型处理复杂的推理任务：

写入 /tmp/deepseek_request.json：

{
  "model": "deepseek-reasoner",
  "messages": [
    {
      "role": "user",
      "content": "What is 15 * 17? Show your work."
    }
  ]
}

curl -s "https://api.deepseek.com/chat/completions" -X POST -H "Content-Type: application/json" -H "Authorization: Bearer $(printenv DEEPSEEK_API_KEY)" -d @/tmp/deepseek_request.json | jq -r '.choices[0].message.content'

推理器模型擅长数学、逻辑和多步骤问题。

5. JSON 输出模式

强制模型返回有效的 JSON：

写入 /tmp/deepseek_request.json：

{
  "model": "deepseek-chat",
  "messages": [
    {
      "role": "system",
      "content": "You are a JSON generator. Always respond with valid JSON."
    },
    {
      "role": "user",
      "content": "List 3 programming languages with their main use cases."
    }
  ],
  "response_format": {
    "type": "json_object"
  }
}

curl -s "https://api.deepseek.com/chat/completions" -X POST -H "Content-Type: application/json" -H "Authorization: Bearer $(printenv DEEPSEEK_API_KEY)" -d @/tmp/deepseek_request.json | jq -r '.choices[0].message.content'

使用消息历史记录继续对话：

写入 /tmp/deepseek_request.json：

{
  "model": "deepseek-chat",
  "messages": [
    {
      "role": "user",
      "content": "My name is Alice."
    },
    {
      "role": "assistant",
      "content": "Nice to meet you, Alice."
    },
    {
      "role": "user",
      "content": "What is my name?"
    }
  ]
}

curl -s "https://api.deepseek.com/chat/completions" -X POST -H "Content-Type: application/json" -H "Authorization: Bearer $(printenv DEEPSEEK_API_KEY)" -d @/tmp/deepseek_request.json | jq -r '.choices[0].message.content'

7. 代码补全（FIM）

使用中间填充进行代码补全（beta 端点）：

写入 /tmp/deepseek_request.json：

{
  "model": "deepseek-chat",
  "prompt": "def add(a, b):\n ",
  "max_tokens": 20
}

curl -s "https://api.deepseek.com/beta/completions" -X POST -H "Content-Type: application/json" -H "Authorization: Bearer $(printenv DEEPSEEK_API_KEY)" -d @/tmp/deepseek_request.json | jq -r '.choices[0].text'

编辑器中的代码补全
填充文档中的空白
上下文感知的文本生成

8. 函数调用（工具）

定义模型可以调用的函数：

写入 /tmp/deepseek_request.json：

{
  "model": "deepseek-chat",
  "messages": [
    {
      "role": "user",
      "content": "What is the weather in Tokyo?"
    }
  ],
  "tools": [
    {
      "type": "function",
      "function": {
        "name": "get_weather",
        "description": "Get the current weather for a location",
        "parameters": {
          "type": "object",
          "properties": {
            "location": {
              "type": "string",
              "description": "The city name"
            }
          },
          "required": ["location"]
        }
      }
    }
  ]
}

curl -s "https://api.deepseek.com/chat/completions" -X POST -H "Content-Type: application/json" -H "Authorization: Bearer $(printenv DEEPSEEK_API_KEY)" -d @/tmp/deepseek_request.json

当模型想要使用函数时，将返回一个 tool_calls 数组。

9. 检查 Token 使用情况

从响应中提取使用信息：

写入 /tmp/deepseek_request.json：

{
  "model": "deepseek-chat",
  "messages": [
    {
      "role": "user",
      "content": "Hello"
    }
  ]
}

curl -s "https://api.deepseek.com/chat/completions" -X POST -H "Content-Type: application/json" -H "Authorization: Bearer $(printenv DEEPSEEK_API_KEY)" -d @/tmp/deepseek_request.json | jq '.usage'

prompt_tokens：输入 token 计数
completion_tokens：输出 token 计数
total_tokens：两者之和

OpenAI SDK 兼容性

DeepSeek 与 OpenAI SDK 完全兼容。只需更改基础 URL：

from openai import OpenAI
client = OpenAI(api_key="your-deepseek-key", base_url="https://api.deepseek.com")

import OpenAI from 'openai';
const client = new OpenAI({ apiKey: 'your-deepseek-key', baseURL: 'https://api.deepseek.com' });

技巧：复杂的 JSON 负载

对于包含嵌套 JSON 的复杂请求（如函数调用），使用临时文件以避免 shell 转义问题：

写入 /tmp/deepseek_request.json：

{
  "model": "deepseek-chat",
  "messages": [{"role": "user", "content": "What is the weather in Tokyo?"}],
  "tools": [{
    "type": "function",
    "function": {
      "name": "get_weather",
      "description": "Get current weather",
      "parameters": {
        "type": "object",
        "properties": {"location": {"type": "string"}},
        "required": ["location"]
      }
    }
  }]
}

curl -s "https://api.deepseek.com/chat/completions" -X POST -H "Content-Type: application/json" -H "Authorization: Bearer $(printenv DEEPSEEK_API_KEY)" -d @/tmp/deepseek_request.json

选择合适的模型：通用任务使用 deepseek-chat，复杂推理使用 deepseek-reasoner
使用缓存：具有相同前缀的重复提示受益于缓存定价（$0.028 vs $0.28）
设置 max_tokens：通过设置适当的限制来防止生成失控
对长响应使用流式传输：对于实时应用有更好的用户体验
JSON 模式需要系统提示：使用 response_format 时，在系统消息中包含 JSON 指令
FIM 使用 beta 端点：代码补全端点在 api.deepseek.com/beta
复杂的 JSON：使用带有 -d @filename 的临时文件以避免 shell 引用问题

🇺🇸English

DeepSeek API

Use the DeepSeek API via direct curl calls to access powerful AI language models for chat, reasoning, and code generation.

Official docs: https://api-docs.deepseek.com/

When to Use

Use this skill when you need to:

Chat completions with DeepSeek-V3.2 model
Deep reasoning tasks using the reasoning model
Code generation and completion (FIM - Fill-in-the-Middle)
OpenAI-compatible API as a cost-effective alternative

Prerequisites

Sign up at DeepSeek Platform and create an account
Go to API Keys and generate a new API key
Top up your balance (no free tier, but very affordable pricing)

export DEEPSEEK_API_KEY="your-api-key"

Pricing (per 1M tokens)

Type	Price
Input (cache hit)	$0.028
Input (cache miss)	$0.28
Output	$0.42

Rate Limits

DeepSeek does not enforce strict rate limits. They will try to serve every request. During high traffic, connections are maintained with keep-alive signals.

How to Use

All examples below assume you have DEEPSEEK_API_KEY set.

The base URL for the DeepSeek API is:

https://api.deepseek.com (recommended)
https://api.deepseek.com/v1 (OpenAI-compatible)

1. Basic Chat Completion

Send a simple chat message:

Write to /tmp/deepseek_request.json:

{
  "model": "deepseek-chat",
  "messages": [
    {
      "role": "system",
      "content": "You are a helpful assistant."
    },
    {
      "role": "user",
      "content": "Hello, who are you?"
    }
  ]
}

Then run:

curl -s "https://api.deepseek.com/chat/completions" -X POST -H "Content-Type: application/json" -H "Authorization: Bearer $(printenv DEEPSEEK_API_KEY)" -d @/tmp/deepseek_request.json

Available models:

deepseek-chat: DeepSeek-V3.2 non-thinking mode (128K context, 8K max output)
deepseek-reasoner: DeepSeek-V3.2 thinking mode (128K context, 64K max output)

2. Chat with Temperature Control

Adjust creativity/randomness with temperature:

Write to /tmp/deepseek_request.json:

{
  "model": "deepseek-chat",
  "messages": [
    {
      "role": "user",
      "content": "Write a short poem about coding."
    }
  ],
  "temperature": 0.7,
  "max_tokens": 200
}

Then run:

curl -s "https://api.deepseek.com/chat/completions" -X POST -H "Content-Type: application/json" -H "Authorization: Bearer $(printenv DEEPSEEK_API_KEY)" -d @/tmp/deepseek_request.json | jq -r '.choices[0].message.content'

Parameters:

temperature (0-2, default 1): Higher = more creative, lower = more deterministic
top_p (0-1, default 1): Nucleus sampling threshold
max_tokens: Maximum tokens to generate

3. Streaming Response

Get real-time token-by-token output:

Write to /tmp/deepseek_request.json:

{
  "model": "deepseek-chat",
  "messages": [
    {
      "role": "user",
      "content": "Explain quantum computing in simple terms."
    }
  ],
  "stream": true
}

Then run:

curl -s "https://api.deepseek.com/chat/completions" -X POST -H "Content-Type: application/json" -H "Authorization: Bearer $(printenv DEEPSEEK_API_KEY)" -d @/tmp/deepseek_request.json

Streaming returns Server-Sent Events (SSE) with delta chunks, ending with data: [DONE].

4. Deep Reasoning (Thinking Mode)

Use the reasoner model for complex reasoning tasks:

Write to /tmp/deepseek_request.json:

{
  "model": "deepseek-reasoner",
  "messages": [
    {
      "role": "user",
      "content": "What is 15 * 17? Show your work."
    }
  ]
}

Then run:

curl -s "https://api.deepseek.com/chat/completions" -X POST -H "Content-Type: application/json" -H "Authorization: Bearer $(printenv DEEPSEEK_API_KEY)" -d @/tmp/deepseek_request.json | jq -r '.choices[0].message.content'

The reasoner model excels at math, logic, and multi-step problems.

5. JSON Output Mode

Force the model to return valid JSON:

Write to /tmp/deepseek_request.json:

{
  "model": "deepseek-chat",
  "messages": [
    {
      "role": "system",
      "content": "You are a JSON generator. Always respond with valid JSON."
    },
    {
      "role": "user",
      "content": "List 3 programming languages with their main use cases."
    }
  ],
  "response_format": {
    "type": "json_object"
  }
}

Then run:

curl -s "https://api.deepseek.com/chat/completions" -X POST -H "Content-Type: application/json" -H "Authorization: Bearer $(printenv DEEPSEEK_API_KEY)" -d @/tmp/deepseek_request.json | jq -r '.choices[0].message.content'

6. Multi-turn Conversation

Continue a conversation with message history:

Write to /tmp/deepseek_request.json:

{
  "model": "deepseek-chat",
  "messages": [
    {
      "role": "user",
      "content": "My name is Alice."
    },
    {
      "role": "assistant",
      "content": "Nice to meet you, Alice."
    },
    {
      "role": "user",
      "content": "What is my name?"
    }
  ]
}

Then run:

curl -s "https://api.deepseek.com/chat/completions" -X POST -H "Content-Type: application/json" -H "Authorization: Bearer $(printenv DEEPSEEK_API_KEY)" -d @/tmp/deepseek_request.json | jq -r '.choices[0].message.content'

7. Code Completion (FIM)

Use Fill-in-the-Middle for code completion (beta endpoint):

Write to /tmp/deepseek_request.json:

{
  "model": "deepseek-chat",
  "prompt": "def add(a, b):\n ",
  "max_tokens": 20
}

Then run:

curl -s "https://api.deepseek.com/beta/completions" -X POST -H "Content-Type: application/json" -H "Authorization: Bearer $(printenv DEEPSEEK_API_KEY)" -d @/tmp/deepseek_request.json | jq -r '.choices[0].text'

FIM is useful for:

Code completion in editors
Filling gaps in documents
Context-aware text generation

8. Function Calling (Tools)

Define functions the model can call:

Write to /tmp/deepseek_request.json:

{
  "model": "deepseek-chat",
  "messages": [
    {
      "role": "user",
      "content": "What is the weather in Tokyo?"
    }
  ],
  "tools": [
    {
      "type": "function",
      "function": {
        "name": "get_weather",
        "description": "Get the current weather for a location",
        "parameters": {
          "type": "object",
          "properties": {
            "location": {
              "type": "string",
              "description": "The city name"
            }
          },
          "required": ["location"]
        }
      }
    }
  ]
}

Then run:

curl -s "https://api.deepseek.com/chat/completions" -X POST -H "Content-Type: application/json" -H "Authorization: Bearer $(printenv DEEPSEEK_API_KEY)" -d @/tmp/deepseek_request.json

The model will return a tool_calls array when it wants to use a function.

9. Check Token Usage

Extract usage information from response:

Write to /tmp/deepseek_request.json:

{
  "model": "deepseek-chat",
  "messages": [
    {
      "role": "user",
      "content": "Hello"
    }
  ]
}

Then run:

curl -s "https://api.deepseek.com/chat/completions" -X POST -H "Content-Type: application/json" -H "Authorization: Bearer $(printenv DEEPSEEK_API_KEY)" -d @/tmp/deepseek_request.json | jq '.usage'

Response includes:

prompt_tokens: Input token count
completion_tokens: Output token count
total_tokens: Sum of both

OpenAI SDK Compatibility

DeepSeek is fully compatible with OpenAI SDKs. Just change the base URL:

Python:

from openai import OpenAI
client = OpenAI(api_key="your-deepseek-key", base_url="https://api.deepseek.com")

Node.js:

import OpenAI from 'openai';
const client = new OpenAI({ apiKey: 'your-deepseek-key', baseURL: 'https://api.deepseek.com' });

Tips: Complex JSON Payloads

For complex requests with nested JSON (like function calling), use a temp file to avoid shell escaping issues:

Write to /tmp/deepseek_request.json:

{
  "model": "deepseek-chat",
  "messages": [{"role": "user", "content": "What is the weather in Tokyo?"}],
  "tools": [{
    "type": "function",
    "function": {
      "name": "get_weather",
      "description": "Get current weather",
      "parameters": {
        "type": "object",
        "properties": {"location": {"type": "string"}},
        "required": ["location"]
      }
    }
  }]
}

Then run:

curl -s "https://api.deepseek.com/chat/completions" -X POST -H "Content-Type: application/json" -H "Authorization: Bearer $(printenv DEEPSEEK_API_KEY)" -d @/tmp/deepseek_request.json

Guidelines

Choose the right model : Use deepseek-chat for general tasks, deepseek-reasoner for complex reasoning
Use caching : Repeated prompts with same prefix benefit from cache pricing ($0.028 vs $0.28)
Set max_tokens : Prevent runaway generation by setting appropriate limits
Use streaming for long responses : Better UX for real-time applications
JSON mode requires system prompt : When using response_format, include JSON instructions in system message
FIM uses beta endpoint : Code completion endpoint is at api.deepseek.com/beta
Complex JSON : Use temp files with -d @filename to avoid shell quoting issues

Weekly Installs

124

Repository

vm0-ai/vm0-skills

GitHub Stars

First Seen

Jan 24, 2026

Security Audits

Gen Agent Trust HubPass SocketPass SnykPass

Installed on

gemini-cli109

opencode108

codex107

github-copilot106

amp102

cursor101

超能力技能使用指南：AI助手技能调用优先级与工作流程详解

48,700 周安装

输入（缓存命中）	$0.028
输入（缓存未命中）	$0.28
输出	$0.42

DeepSeek API 使用指南：聊天、推理与代码生成，OpenAI 经济替代方案

🇨🇳中文介绍

DeepSeek API

使用场景

先决条件

定价（每 100 万 token）

相关 Skills

速率限制

使用方法

1. 基础聊天补全

2. 带温度控制的聊天

3. 流式响应

4. 深度推理（思考模式）

5. JSON 输出模式

6. 多轮对话

7. 代码补全（FIM）

8. 函数调用（工具）

9. 检查 Token 使用情况

OpenAI SDK 兼容性

技巧：复杂的 JSON 负载

指南

🇺🇸English

DeepSeek API

When to Use

Prerequisites

Pricing (per 1M tokens)

Rate Limits

How to Use

1. Basic Chat Completion

2. Chat with Temperature Control

3. Streaming Response

4. Deep Reasoning (Thinking Mode)

5. JSON Output Mode

6. Multi-turn Conversation

7. Code Completion (FIM)

8. Function Calling (Tools)

9. Check Token Usage

OpenAI SDK Compatibility

Tips: Complex JSON Payloads

Guidelines

最新 Skills