Moonshot AI's flagship open-weight model, with 2.8 trillion parameters built on Kimi Delta Attention and Attention Residuals. It has native visual understanding and is designed for long-horizon coding, knowledge work and deep reasoning.
- Modalities
- text, image, video → text
- In / out price30% off
List price $3 / $15$2.093511 / $10.467555per 1M tokens- Context
- 1Mtokens
- Released
- 2026-07-16
Pricing
Prices in US dollars, charged from your prepaid balance as you go. No subscription.
30% off30% below the list price, on every token price.
- Input
List price $3$2.093511/ 1M tokens- Output
List price $15$10.467555/ 1M tokens- Cache read
List price $0.30$0.209351/ 1M tokens- Cache write
List price $3$2.093511/ 1M tokens
Specs
Limits and capabilities as published by the model’s provider.
- Context
- 1Mtokens
- Max output
- 1Mtokens
- Input types
- text, image, video
- Output types
- text
- Released
- 2026-07-16
- Open weights
- Hugging Face
- License
- Kimi K3 License
Call this model
The examples read your key from the CHEAPAIMODEL_API_KEY environment variable.
curl https://api.cheapaimodel.ai/v1/chat/completions \
-H "Authorization: Bearer $CHEAPAIMODEL_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "kimi-k3",
"messages": [{"role": "user", "content": "Hello!"}]
}'Endpoints
Endpoints that accept this model, each with its API reference.
- API reference
POST /v1/chat/completionsChat Completions · OpenAI compatible - API reference
POST /v1/responsesResponses · OpenAI compatible
Compare with other models
Kimi K3 side by side with each other model in the catalog: price, context and the same code.
From the blog
Posts on our blog that cover Kimi K3: pricing, setup and troubleshooting.
- Kimi API International: Payment, Sign-Up and Rate Limits
Using the Kimi API outside mainland China: which platform to pick, how sign-up and top-ups work, the rate-limit tiers, and a card-paid route to Kimi K3.
- Kimi API Key and Pricing: Set Up K3 and Estimate Your Costs
Get the right Kimi API key, compare access routes, understand K3 token and cache charges, and check your coding-tool configuration before scaling.
- Kimi vs DeepSeek API Cost: Long Context, Output and Caching
Compare Kimi K3 and DeepSeek V4.1 Flash API costs with long-document examples, cache billing differences, and DeepSeek peak versus off-peak rates.