Official2026-06-11
DeepSeek Rate Limits Use Account Concurrency and user_id
Checked on June 11, 2026: DeepSeek's official docs now give a cleaner production picture for V4 traffic management, with 500 Pro concurrency, 2500 Flash concurrency, explicit `user_id` isolation rules, and request keep-alive handling.
What we verified on June 11, 2026
- DeepSeek's official Rate Limit & Isolation page sets account-level concurrency at 500 for
deepseek-v4-proand 2500 fordeepseek-v4-flash. - The same official page now explains what
user_iddoes: content-safety isolation, KV-cache isolation, and scheduling isolation under one account. - DeepSeek also documents protocol-specific
user_idplacement:extra_body.user_idfor OpenAI-format calls andmetadata.user_idfor Anthropic-format calls. - The same page documents the request keep-alive behavior: non-streaming responses can emit empty lines, streaming responses can emit SSE keep-alive comments, and the server closes the connection if inference has not started after 10 minutes.
- DeepSeek's English homepage still points its public social anchor to
@deepseek_ai, which remains the safest official X confirmation even though the X page itself is not safely crawlable here.