DeepSeek API pricing
DeepSeek API pricing: current per-token rates
DeepSeek's current API rates per 1M tokens, input / output / cached-read, verified 2026-09-06. Every rate links to its official source.
Current DeepSeek model prices
| Model | Input ($/1M) | Output ($/1M) | Cached in ($/1M) | Context |
|---|---|---|---|---|
| DeepSeek V3.1 | $0.55 | $1.65 | $0.55 | 164K |
| DeepSeek V3.2 | $0.27 | $0.40 | $0.13 | 164K |
| DeepSeek V4 Flash | $0.08 | $0.16 | $0.02 | 1M |
| DeepSeek V4 Pro | $0.74 | $1.48 | $0.06 | 1M |
Non-official data by OpenRouter (passthrough pricing — the provider’s own rate, no markup); verify against the provider's own page before relying on it.
Recent price changes
- DeepSeek V4 Flash: $0.09/$0.17 → $0.08/$0.16 per 1M (effective 2026-09-06)
- DeepSeek V4 Pro: $0.94/$1.88 → $0.74/$1.48 per 1M (effective 2026-09-06)
- DeepSeek V3.1: $0.25/$0.95 → $0.55/$1.65 per 1M (effective 2026-09-04)
- DeepSeek V3.2: $0.26/$0.38 → $0.27/$0.40 per 1M (effective 2026-08-30)
Common questions
Can I use the DeepSeek API for free?
No — API access is metered per token at the rates above. Subscriptions (where offered) bundle usage into a flat monthly price; the API itself is pay-as-you-go.
Which model is cheapest?
The lowest input/output rate in the table above for light work; match the model to the task rather than defaulting to the cheapest — routing the hard steps up and the cheap steps down usually beats a single choice.
Prices change. Verified 2026-09-06 — re-check the source before relying on them. Corrections: hello@aiarch.dev.
Learn to cost-model agents like an architect.
What a Claude agent actually costs, or compare Professional Membership pricing.