Source: Models and pricing (DeepSeek API docs)
DeepSeek's API models (V4.1-Flash, V4-Pro): 1M context, thinking modes, cache-hit pricing and off-peak discounts.
Contents
- What: official API documentation page
- Publisher: DeepSeek
- Date: accessed 2026-11-14
- Original: Models & Pricing
Summary
Describes the two API models, which map to the open-weight DeepSeek-V4.1-Flash-1017 and DeepSeek-V4-Pro-0813 checkpoints, and their pricing.
Key claims
deepseek-flashserves DeepSeek-V4.1-Flash-1017;deepseek-v4-proserves DeepSeek-V4-Pro-0813; both 1M context with thinking (default) and non-thinking modes.- Peak prices: Flash $0.30 input (cache miss) / $1.20 output; V4-Pro $1.32 / $3.96. Off-peak is half.
- Cache-hit input is 50x cheaper than cache-miss input on Flash ($0.006 vs $0.30) and 30x cheaper on V4-Pro ($0.044 vs $1.32).
- Peak hours are 01:00-04:00 and 06:00-10:00 UTC on weekdays (excluding Chinese public holidays); all other times are off-peak.
- Max output is listed as 384K tokens.
- The older
deepseek-v4-flashanddeepseek-v4-flash-vision-expnames are retired but still accepted; requests are served by V4.1-Flash-1017 at the Flash price. - Both expose OpenAI-compatible and Anthropic-compatible endpoints, JSON output and tool calls.
- Flash supports vision; V4-Pro does not.