DeepSeek API
The DeepSeek API provides programmatic access to DeepSeek's language models, with half-price rates outside peak hours.
Usage-basedVerified
Our verdict
Very cheap when your prompts repeat: cache-hit input costs a fraction of a cent per million tokens, and off-peak rates are half price. On fresh, uncached prompts its Flash model costs more than the smallest OpenAI, Anthropic and Gemini models.
Best for
- ✓ Apps that reuse the same long prompt
- ✓ Batch jobs you can run off-peak
Watch out for
- ! Peak hours (weekdays 01:00-04:00 and 06:00-10:00 UTC) cost double
- ! Search and Fetch cost $0.75 per 1,000 requests; calls made through DeepSeek Harness, Claude Code or Codex start being billed on November 1, 2026
Pricing
Read from the official pricing page. Prices in USD; always confirm on the vendor's site.
| Plan | Price | Details |
|---|---|---|
| deepseek-flash (V4.1 Flash) | Pay as you go | Peak: $0.006 per 1M cache-hit input, $0.30 per 1M cache-miss input, $1.20 per 1M output. Off-peak: $0.003 / $0.15 / $0.60 |
| deepseek-v4-pro | Pay as you go | Peak: $0.044 per 1M cache-hit input, $1.32 per 1M cache-miss input, $3.96 per 1M output. Off-peak: $0.022 / $0.66 / $1.98 |
Note: USD table on the official Models & Pricing page. deepseek-flash is served by DeepSeek-V4.1-Flash-1017. Off-peak rates are half of peak; peak hours are 01:00-04:00 and 06:00-10:00 UTC Monday to Friday, excluding Chinese public holidays.