Connecting to model catalogs and market prices…
Moonshot AI
Kimi K3
Verified suppliersTrusted suppliers sell this model through GPU Router, and our marketplace orders only go to sellers vetted as trusted.moonshotai/kimi-k3Est. 81% off · Standardvs cheapest OpenRouter endpoint for a typical request: ~128k input / 1k outputKimi K3 is a 2.8T parameter open-weight multimodal reasoning model from Moonshot AI. It is suited for complex coding, knowledge work, and long-horizon agentic workflows, and is particularly strong at...
Standard prices and discounts apply to Standard requests. Requests requiring zero data retention use separate ZDR quotes. Retention depends on the provider serving your request.
Marketplace price history.
Marketplace observations and reference prices.
Higher is better. Each point prices a typical request of the model (~128k input / 1k output) at the recorded market rates, against today’s cheapest OpenRouter endpoint for the same request, using input and output rates only on both sides; OpenRouter history is not recorded. Negative values mean above the OpenRouter reference. Dashed lines show current references and the gateway estimate. Recorded market prices exclude the routing fee; gaps stay visible. In + out = 1M input + 1M output tokens.
Use this model
Kimi K3 moonshotai/kimi-k3
- 01Set up GPU Router
Your coding assistant adds the client and checks model availability.
- 02Choose your calls
Switch all matching calls, selected features, or use it for new calls only.
- 03Switch and test
Only the calls you choose are migrated. Everything else stays as is.
Paste into your coding assistant. It will ask before switching calls.
What would your workload cost?
Change your token mix to compare the same workload.
For 128M input + 1M output tokens using our current estimate.
Our price is the cheapest eligible route or live trusted seller offer for this workload, including the routing fee. Final charges depend on the route and actual usage.
