Home / DeepSeek API pricing
DeepSeek API pricing
What DeepSeek charges for its models through the API, per million tokens. Prices change with the time of day: off-peak hours cost half.
Checked October 13, 2026 against DeepSeek’s models and pricing page
Your monthly DeepSeek cost
Models
- Turn on JavaScript to see the costs. The price tables below work without it.
Prices
| Model | Cache hit | Cache miss (input) | Output |
|---|---|---|---|
| Peak: 01:00–04:00 and 06:00–10:00 UTC, Monday to Friday | |||
| DeepSeek Flashdeepseek-flash · DeepSeek-V4.1-Flash | $0.006 | $0.30 | $1.20 |
| DeepSeek V4 Prodeepseek-v4-pro · DeepSeek-V4-Pro-0813 | $0.044 | $1.32 | $3.96 |
| Off-peak: all other hours, weekends and Chinese public holidays | |||
| DeepSeek Flash | $0.003 | $0.15 | $0.60 |
| DeepSeek V4 Pro | $0.022 | $0.66 | $1.98 |
What changes the bill
Time of day
Off-peak rates are half the peak rates. Peak hours are 01:00–04:00 and 06:00–10:00 UTC on weekdays, which is 9 a.m. to noon and 2 p.m. to 6 p.m. in Beijing. Chinese public holidays are off-peak all day. Work you can schedule outside those hours costs half.
Cache hits
Input DeepSeek has already seen and cached costs far less: $0.006 instead of $0.30 per million on Flash at peak, a fiftieth of the price.
Old model names
DeepSeek still accepts the older names deepseek-v4-flash and deepseek-v4-flash-vision-exp, but those models are retired. Requests are answered by DeepSeek-V4.1-Flash and billed at the Flash price.
Paying
DeepSeek takes the cost from a balance you top up in advance, using any granted balance first. Its pricing page lists no batch price.