DeepSeek V4 Pro API 上线：1M 上下文、384K 最大输出

``` curl \curl \ -H "Content-Type: application/json" \ -H "Content-Type: application/json" \ -H "Authorization: Bearer ${DEEPSEEK_API_KEY}" \ -H "Authorization: Bearer ${DEEPSEEK_API_KEY}" \ -d '{ -d '{ "model": "deepseek-v4-pro", "model": "deepseek-v4-pro", "messages": [ "messages": [ {"role": "system", "content": "You are a helpful assistant."}, {"role": "system", "content": "You are a helpful assistant."}, {"role": "user", "content": "Hello!"} {"role": "user", "content": "Hello!"} ], ], "thinking": {"type": "enabled"}, "thinking": {"type": "enabled"}, "reasoning_effort": "high", "reasoning_effort": "high", "stream": false "stream": false }' }' ``` [...] # Please install OpenAI SDK first: `pip3 install openai` import os import os from openai import OpenAI from openai import OpenAI client = OpenAI(client = OpenAI( api_key=os.environ.get('DEEP

Skip to main content Image 1: DeepSeek API Docs Logo DeepSeek API Docs English English 中文（中国） DeepSeek Platform Quick Start Your First API Call Models & Pricing Token & Token Usage Rate Limit & Isolation Error Codes Agent Integrations API Guides Thinking Mode Multi-round Conversation Chat Prefix Completion (Beta) FIM Completion (Beta) JSON Output Tool Calls Context Caching Using the Responses API Using the Anthropic API API Reference News Other Resources FAQ Change Log BASE URL (Anthropic Format) MODEL VERSION DeepSeek-V4-Flash-0731 DeepSeek-V4-Pro THINKING MODE Supports both non-thinking and thinking (default) modes [...] See Thinking Mode for how to switch CONTEXT LENGTH 1M MAX OUTPUT MAXIMUM: 384K FEATURESJson Output✓✓ Tool Calls✓✓ Responses API(1)✓✗ Anthropic API✓✓ Chat Prefix Completion（Beta）✓✓ FIM Completion（Beta）Non-thinking mode on

Note: This update only upgrades the DeepSeek-V4-Flash API. The DeepSeek-V4-Pro API and the APP/WEB models are unchanged. The official release of DeepSeek-V4-Pro will follow soon. ## Date: 2026-04-24​ ### DeepSeek-V4​ The DeepSeek API now supports V4-Pro and V4-Flash, available via both the OpenAI ChatCompletions interface and the Anthropic interface. To access the new models, the base_url remains unchanged, and the model parameter should be set to `deepseek-v4-pro` or `deepseek-v4-flash`. The two legacy API model names, `deepseek-chat` and `deepseek-reasoner`, will be discontinued in three months (2026-07-24). During the current period, these two model names point to the non-thinking mode and thinking mode of `deepseek-v4-flash`, respectively. [...] Skip to main content Image 1: DeepSeek API Docs Logo DeepSeek API Docs English English 中文（中国） DeepSeek Platform Quick Start Your First API Call Models & Pricing Token & Token Usage Rate Lim

"reasoning_summary_format": "experimental", "default_reasoning_summary": "none", "display_name": "DeepSeek-V4-Pro", "description": "Most capable frontier agentic coding model.", "default_reasoning_level": "high", "supported_reasoning_levels": [ { "effort": "low", "description": "Fast responses with lighter reasoning" }, { "effort": "high", "description": "Extra high reasoning depth for complex problems" }, { "effort": "max", "description": "Maximum reasoning depth for the hardest problems" } ], "shell_type": "shell_command", "visibility": "list", "minimal_client_version": "0.144.0", "supported_in_api": true, "availability_nux": null, "upgrade": null, "priority": 2, "model_messages": { [...] DeepSeek API 文档 Logo DeepSeek API 文档 Logo # 接入 Codex 目前仅 `deepseek-v4-flash` 支持接入 Codex，`deepseek-v4-pro` 预计 2026 年 8 月初支持接入 Codex。 `deepseek-v4-flash` `deepseek-v4-pro` Codex 是 OpenAI 推出的 AI 编程助手，通过 Responses API 与模型交

跳到主要内容 Image 1: DeepSeek API 文档 Logo DeepSeek API 文档 中文（中国） English 中文（中国） DeepSeek Platform 本页总览 # 集成 Deep Code Deep Code 是一款开源的终端 AI 编程助手，专为 DeepSeek-V4 系列模型适配，支持深度思考、推理强度控制以及 Agent Skills。 GitHub： #### 1. 安装 Deep Code​ 安装 Node.js 18+ 版本。 在终端中运行以下命令： `npm install -g @vegamo/deepcode-cli` 验证安装是否成功： `deepcode --version` #### 2. 配置 Deep Code​ 创建 `~/.deepcode/settings.json` 文件，填入你的 DeepSeek API Key 和模型配置： `{ "env": { "MODEL": "deepseek-v4-pro", "BASE_URL": " "API_KEY": "sk-..." }, "thinkingEnabled": true, "reasoningEffort": "max"}` 从 DeepSeek 开放平台 获取你的 API Key。 > 注意： 此配置文件与 Deep Code VSCode 扩展 共享。 配置选项：

Official Documentation: ## DeepSeek V4 ProOverview | | | --- | | Model Name | DeepSeek V4 Pro | | Provider | DeepSeek | | Model ID | `deepseek-v4-pro` | | Release Date | Apr 24, 2026 | | Last Updated | Apr 24, 2026 | | Knowledge Cutoff | 2025-05 | | Context Window | 1,000,000 tokens | | Max Output | 384,000 tokens | | Pricing /1M tokens | $0.435 input $0.87 output $0.87 reasoning $0.004 cache read | | Input Modalities | text | | Output Modalities | text | | Capabilities | ReasoningTool CallingTemperature ControlOpen Weights | ## Complete Setup Guide 1 ### Get Your DeepSeekAPI Key First, you'll need to obtain an API key fromDeepSeek. This key allows you to access their AI models directly and pay only for what you use. [...] DeepSeek AI is a Chinese open-source AI company offering advanced large language models with the latest DeepSeek-V3.1 (released August 2025) combining both general-purpose and reasoning capabilities in a hybrid architecture. Key models include DeepSeek-V3.1