Fresh paint, wet floors — we're rebuilding CostOfToken into a cross-provider price comparison. Some corners are still under construction; the full version is coming up soon.

Kimi API pricing

Every Kimi model from Moonshot AI we track, with input, cached input and output cost in USD per 1M tokens. Moonshot’s Kimi models target long-context work at a fraction of Western flagship pricing.

8 models · updated 2026-09-25 · prices in USD per 1,000,000 tokens

8
Models tracked
$0.450
Cheapest input — Kimi K2.5
$0.950
Priciest input — Kimi K2.6
1M
Largest context — Kimi K3

Listed in the order Moonshot AI (Kimi) publishes them.

Kimi models with input, cached input and output pricing per million tokens
Kimi K3Provider API$0.884$0.330$10.531M
Kimi K2.7 CodeProvider API$0.656$0.180$3.30262K
Kimi LatestProvider API$0.884$0.330$10.531M
Kimi K2.6Provider API$0.950$0.160$4.00262K
Kimi K2.5Provider API$0.450$0.070$2.25262K
Kimi K2 ThinkingProvider API$0.600$0.150$2.50262K
Kimi K2 0905Provider API$0.600—$2.50262K
Kimi K2 0711Provider API$0.570—$2.30131K

Kimi pricing questions

How much does the Kimi API cost?
Kimi pricing runs from $0.450 to $0.950 per 1M input tokens depending on the model. Kimi K2.6 costs $0.950 per 1M input and $4.00 per 1M output tokens.
What is the cheapest Kimi model?
Kimi K2.5 is the cheapest Kimi model we track, at $0.450 per 1M input tokens and $2.25 per 1M output tokens.
Does Kimi charge less for cached tokens?
Yes. Cached input is billed at a lower rate than fresh input — for Kimi K3 it is $0.330 per 1M versus $0.884, about 63% cheaper. Caching applies to repeated prompt prefixes.
How often is this Kimi pricing updated?
Every day. Prices are read directly from the published pricing page, normalized to USD per 1,000,000 tokens, and only recorded when a number actually changes — so the history shows real price movements rather than daily noise.

Compare with other providers