Back to approved models
Kimi AIactiverelay/request.mjsready to route CONTEXT1M contextprompt window MAX OUTPUT1Mper response INPUT$3.0 / Mtoken price OUTPUT$15.0 / Mtoken price
Approved model profile_
Kimi K3
Kimi 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...
kimi-k3const completion = await client.chat.completions.create({
model: "kimi-k3",
messages: [
{ role: "user",
content: "Plan a multi-step task" }
]
});Route configuration_
Use the model through the contract you already have.
PipeLLM resolves provider access, policy, and supported model capabilities without changing the interface your application calls.
https://api.pipellm.aiOne base URL for OpenAI, Anthropic, and Gemini.FORMAT ROUTES
OPENAI
/openaiANTHROPIC
/anthropicGEMINI
Use a format route to normalize requests and responses./geminikimi-k3Use this in every requestPROVIDER AVAILABILITY2 routes
| Provider | Context | Max output | Discount | Input / M | Output / M |
|---|---|---|---|---|---|
| Kimi AI | 1M context | 1M | — | $3.0 / M | $15.0 / M |
| Tokenhub | 1M context | 1M | — | $3.0 / M | $15.0 / M |
SUPPORTED SURFACESactive
input: textoutput: texttool usecache controlcomputer use
Relay governs model access with the same routing policy that Lens keeps visible and auditable.
