Kimi K3 API pricing

What is Kimi K3?

Kimi K3 is a text model in the Outrun catalog. Its developer describes it for long-horizon coding, knowledge work and reasoning. The standard rate is $1.80 in / $9.00 out per 1M tokens.

What are the Kimi K3 model details?

How do I call Kimi K3?

Use your Outrun key and the exact model ID. Requests use prepaid USD credit.

curl https://api.outrun.to/v1/chat/completions \
  -H "Authorization: Bearer $OUTRUN_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
  "model": "kimi-k3",
  "messages": [
    {
      "role": "user",
      "content": "Hello!"
    }
  ],
  "max_tokens": 8192
}'

import os
from openai import OpenAI

client = OpenAI(
    base_url="https://api.outrun.to/v1",
    api_key=os.environ["OUTRUN_API_KEY"],
)
reply = client.chat.completions.create(
    model="kimi-k3",
    messages=[{"role": "user", "content": "Hello!"}],
    max_tokens=8192,
)
print(reply.choices[0].message.content)

Which tools work with Kimi K3?

Any client that accepts an OpenAI-compatible base URL and a model ID can use kimi-k3. Setup guides: Cline (OpenAI Compatible), OpenCode, SillyTavern, LiteLLM.

Use these setup values for this model ID. Recommended output settings can be lower than the route maximum.

How does Kimi K3 compare on cost?

Assumes 1,000 input + 300 output tokens per turn, repeated 1,000 times, at standard-tier rates without caching. Input / output shown per 1M tokens, rounded up to the cent.

Which similar models can I try?

Common questions about Kimi K3

Where can I find more models?