DeepSeek API pricing

DeepSeek API pricing: current per-token rates

Short answer

DeepSeek's current API rates per 1M tokens, input / output / cached-read, verified 2026-10-19. Every rate links to its official source.

Current DeepSeek model prices

ModelInput ($/1M)Output ($/1M)Cached in ($/1M)Context
DeepSeek V3.1$0.25$0.95$0.13164K
DeepSeek V3.1 Terminus$0.27$1—164K
DeepSeek V3.2$0.28$0.42$0.03164K
DeepSeek V4 Flash$0.09$0.18$0.021M
DeepSeek V4 Pro$0.96$1.91$0.081M
DeepSeek V4.1 Flash$0.00$0.60$0.011M

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 Pro: $0.21/$0.42 → $0.96/$1.91 per 1M (effective 2026-10-19)
  • DeepSeek V3.2: $0.26/$0.80 → $0.28/$0.42 per 1M (effective 2026-10-17)
  • DeepSeek V4 Flash: $0.03/$0.06 → $0.09/$0.18 per 1M (effective 2026-10-14)
  • DeepSeek V4.1 Flash: $0.00/$2.40 → $0.00/$0.60 per 1M (effective 2026-10-14)
  • DeepSeek V3.1 Terminus: $0.30/$1 → $0.27/$1 per 1M (effective 2026-10-03)
  • DeepSeek V3.1: $0.55/$1.65 → $0.25/$0.95 per 1M (effective 2026-09-08)

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.

Sources & provenance

Prices change. Verified 2026-10-19 — re-check the source before relying on them. Corrections: hello@aiarch.dev.

Learn to cost-model agents like an architect.

Free with an account: Track 0 (The Mindset Shift) and the Track 0 coach allowance, no card · every claim cited · full curriculum with membership