What is the DeepSeek API base URL?
DeepSeek is OpenAI-compatible. Keep the standard openai library and change only base_url, api_key, and the model name.
base_url="https://api.deepseek.com/v1"Python quickstart guideSource: catalog-fallback
DeepSeek API documentation, organized for your first call
Start with the DeepSeek quick answers below, then browse the official documentation links. The directory refresh time reflects link metadata retrieval, not the date every source or fact was verified.
Providers
7
Pages
28
Directory refreshed
Nov 6, 2026, 11:37 AM
DeepSeek API quick answers
The five things developers ask before their first call. Each answer links to a full guide.
Which DeepSeek model ID should I use?
Use deepseek-flash (DeepSeek-V4.1-Flash) as the default for chat, streaming, and repeated tool steps. Switch to deepseek-v4-pro for harder reasoning or review-heavy work. Both have a 1M-token context window.
Pro vs Flash guideHow do thinking mode and tool calls work?
Thinking mode returns a reasoning_content field alongside the answer, and DeepSeek supports OpenAI-style function calling / tools. Use Flash for high-frequency tool loops and Pro for harder chains.
Thinking & tool-call guideWhat does the DeepSeek API cost?
Per 1M tokens, cache-hit / cache-miss / output: DeepSeek V4.1 Flash off-peak: $0.003 / ¥0.02 / $0.15 / ¥1.00 / $0.6 / ¥4.00; DeepSeek V4.1 Flash peak: $0.006 / ¥0.04 / $0.3 / ¥2.00 / $1.2 / ¥8.00; DeepSeek V4 Pro off-peak: $0.022 / ¥0.15 / $0.66 / ¥4.50 / $1.98 / ¥13.50; DeepSeek V4 Pro peak: $0.044 / ¥0.30 / $1.32 / ¥9.00 / $3.96 / ¥27.00. Peak hours are 09:00–12:00 and 14:00–18:00 Beijing time (01:00–04:00 and 06:00–10:00 UTC), Monday–Friday, excluding Chinese public holidays. All other hours, weekends, and Chinese public holidays are off-peak. Verified 2026-10-11; official API rates are separate from Coding Plans.
API pricing guideOpenAI-compatible or Anthropic-compatible endpoint?
Use the /v1 OpenAI-compatible endpoint for standard SDKs, or DeepSeek's Anthropic-compatible endpoint for Claude Code. Use deepseek-v4-pro or deepseek-flash. The old deepseek-v4-flash and deepseek-v4-flash-vision-exp aliases now use V4.1-Flash and its prices.
OpenAI vs Anthropic routing guideD
DeepSeek
DeepSeek
DeepSeek official docs cover first API calls, model and pricing details, thinking mode, tool usage, and the API reference.
DeepSeek-V4-ProDeepSeek-V4-FlashDeepSeek-V3.2 legacy aliases
quickstartCatalog fallback
Your First API Call | DeepSeek API Docs
Start from the API key and base URL, then send your first chat/completions request.
Source
api-docs.deepseek.com
modelsCatalog fallback
Models & Pricing
Review DeepSeek V4 Pro and Flash model routes, current context limits, and pricing guidance.
Source
api-docs.deepseek.com
referenceCatalog fallback
API Reference
Entry point for DeepSeek authentication, endpoints, and API reference material.
Source
api-docs.deepseek.com
guidesCatalog fallback
Thinking Mode
Explains reasoning_content, thinking parameters, and compatibility notes for thinking mode.
Source
api-docs.deepseek.com