DeepSeek

What is DeepSeek?

DeepSeek's chat and API with open-weight reasoning models.

Who is DeepSeek best for?

DeepSeek is best for developers who want open-weight models with long context.

• High-volume API work

  Flash costs $0.3 per million input tokens at peak and half that off-peak, and cached input costs a fraction of a cent per million.

• Long documents and agent runs

  The API takes a 1M-token context and returns up to 384K tokens, with tool calls and JSON output.

• Moving code off another provider

  OpenAI-format and Anthropic-format endpoints let existing code and agents such as Claude Code or Codex run on DeepSeek models.

• Running models yourself

  DeepSeek publishes its model weights, including V4.1-Flash on Hugging Face, for teams that want to host models on their own hardware.

How much does DeepSeek cost?

DeepSeek publishes prices only for its pay-as-you-go API, where the deepseek-flash model costs $0.3 per million input tokens and $1.2 per million output tokens at peak times, and half that off-peak.

Off-peak rates are half the peak rates. Peak hours are 01:00 to 04:00 and 06:00 to 10:00 UTC, Monday to Friday; all other hours, weekends and Chinese public holidays are off-peak. The pricing page still lists deepseek-v4-pro at $1.32 / $3.96 per million input / output tokens at peak, but DeepSeek now serves those calls with V4.1-Flash at Flash rates until V4.1-Pro launches. You pay from a topped-up balance (PayPal, card, Alipay or WeChat Pay) that does not expire, and unused balance can be refunded. DeepSeek lists no price or paid plan for its web chat and app. Full pricing on api-docs.deepseek.com. We checked these prices on 27 Sep 2026.

How do readers rate DeepSeek?

No one has rated DeepSeek yet. Be the first.

Questions about DeepSeek