Skip to content
NewTry RouterPlus with free creditsTry playground →
RouterPlus
Book a demo

Kimi API

Open in consoleI need inference

Pricing

ModelReleasedContextInput / 1MOutput / 1MProvidersWeights
Kimi K3moonshotai/kimi-k3July 16, 20261.05M$1.50Sail Research$10.00Sail Research6Open
Kimi K3 Fastmoonshotai/kimi-k3-fastJuly 16, 20261.05M$4.50Inference.net$22.50Inference.net1Open
Kimi K2.6moonshotai/kimi-k2.6April 20, 2026262K$0.95Nebius$4.00Nebius3Open
Kimi K2.5moonshotai/kimi-k2.5January 27, 2026262K$0.80Inference.net$3.95Inference.net1Open

Lowest listed price, USD per 1M tokens, as of October 27, 2026.

FAQ

Which Kimi models can I use through RouterPlus?

Kimi K3, Kimi K3 Fast, Kimi K2.6 and Kimi K2.5. Each one uses its catalog ID, such as "moonshotai/kimi-k3", with the OpenAI-compatible endpoint at https://api.routerplus.com/v1.

What is the cheapest Kimi model on RouterPlus?

Kimi K2.5, at $0.80 per million input tokens and $3.95 per million output tokens on Inference.net. Prices as of October 27, 2026.

Which Kimi model has the most providers?

Kimi K3, served by 6 providers: Morph, DeepInfra, Nebius, Novita AI, Inference.net and Sail Research.

What is the newest Kimi model?

Kimi K3, released July 16, 2026, is the newest Moonshot AI model in the RouterPlus catalog.

Run Kimi in production

Open in consoleI need inference