DeepSeek API review

The DeepSeek API ranks seventh with the highest value score on the list. Its pricing doc lists two models with a 1M-token context: deepseek-flash, now served by DeepSeek-V4.1-Flash-1017, at $0.30 input and $1.20 output per 1M tokens at peak, and deepseek-v4-pro at $1.32 and $3.96. Cache hits on Flash cost $0.006 per 1M tokens, and every rate halves in off-peak hours, which cover weekends and all weekday hours outside the two UTC peak windows. The API accepts both OpenAI and Anthropic request formats, so tools built for either SDK can switch with a base URL change, and both models support JSON output, tool calls and the Responses API. The limits are scope and options: there are only two models, Pro has no vision input, and there is less enterprise packaging than on Azure or Bedrock.

Who it is for

Cost-focused developers who want a 1M-token context model and can schedule work into off-peak hours.

Price

Pay per token; deepseek-flash $0.30 per 1M input tokens at peak, half off-peak.

1M INPUT TOKENS (CACHE MISS) OFF-PEAK $0.15 $0.66 PEAK $0.3 $1.32
Quoted from the vendor pricing page on November 7, 2026.

Key features

• Two models: deepseek-flash, served by DeepSeek-V4.1-Flash-1017, and deepseek-v4-pro
• 1M-token context length
• Base URLs for both OpenAI and Anthropic API formats
• Off-peak rates at half the peak rates
• JSON output, tool calls and the Responses API on both models

Strengths and weaknesses

• Strength: Cache hits on deepseek-flash cost $0.006 per 1M tokens at peak
• Strength: Off-peak hours halve every token price
• Strength: Works with the OpenAI and Anthropic SDKs by changing the base URL
• Weakness: Only two models, and vision is not supported on deepseek-v4-pro
• Weakness: Peak hours are weekday UTC windows, so savings depend on timing
• Weakness: Fewer enterprise deployment options than the cloud platforms here

Key facts

• deepseek-flash: $0.30 input and $1.20 output per 1M tokens at peak
• deepseek-v4-pro: $1.32 input and $3.96 output per 1M tokens at peak
• Context: 1M tokens

Where it ranks

DeepSeek API is No. 7 in our Best LLM APIs verdict.