DeepSeek

1. Your first API call

   OpenAI-compatible endpoint, API keys, and base URL.

2. Models and pricing

   deepseek-chat and deepseek-reasoner models, token pricing, and cache pricing.

3. Parameter settings

   Temperature and sampling recommendations by use case.

4. Multi-round chat

   Managing conversation history in a stateless API.

5. Reasoning model

   Chain-of-thought output, reasoning_content, and usage constraints.

6. JSON output

   Forcing valid JSON responses.

7. Tool calls

   Defining tools and handling model tool calls.

8. Chat prefix and FIM completion

   Prefix completion and fill-in-the-middle for code.

9. Context caching

   Disk-based KV cache hits that reduce cost and latency.

10. Token usage

    How tokens are counted and billed.

11. Rate limit

    Dynamic concurrency limits and handling congestion.

12. Error codes

    Error codes and recommended resolutions.

Each section links to the official page. Titles and summaries by Extra Credit; the docs belong to their authors.