Context
1.0M
Max output
1.0M
Input / 1M
$3.00
Output / 1M
$15.00
Cached / 1M
$0.30
Cutoff
Jan 2026
About this model
Kimi K3 is Moonshot AI's most capable flagship model with 2.8 trillion parameters, native visual understanding, and a 1M-token context window. Designed for frontier intelligence scenarios including software engineering, knowledge work, and deep reasoning.
Best suited for
- Frontier intelligence and complex reasoning
- Software engineering and coding
- Long-context document analysis (1M tokens)
- Multimodal tasks with image and video understanding
- Knowledge work and research
- Agentic workflows
Capabilities
Vision
Accepts images alongside text in the same message.
Tools
Native function calling, so agents can invoke your endpoints.
System prompt
Honours a dedicated system role, separate from the user turn.
Supported parameters
max_tokensSpecifies the maximum number of tokens allowed in the response.
toolsLists tool definitions or capabilities available to the model.
tool_choiceDecides whether to use tools or just the model for generating responses.
response_typeDefines the format of the generated response.
streamSends the response in real-time as it is being generated.
Use this model
OpenAI-SDK compatible, with gateway fallback and routing across providers.
Other Moonshot models
| Model | Context | Max output | Input | Output | Capabilities |
|---|---|---|---|---|---|
| kimi-k3.1-pro-highspeed Chat | 1.0M | — | $4.00 | $20.00 | Vision, Tools, System prompt |
| kimi-k3.1-pro Chat | 1.0M | — | $2.00 | $10.00 | Vision, Tools, System prompt |
| kimi-k2.7-code-highspeed Chat | 262K | 262K | $1.90 | $8.00 | Vision, Files, Tools, System prompt |
| kimi-k2.7-code Chat | 262K | 262K | $0.95 | $4.00 | Vision, Files, Tools, System prompt |
| kimi-k2.6 Chat | 262K | 262K | $0.95 | $4.00 | Vision, Files, Tools, System prompt |
Pricing & provider details
Moonshot
moonshot/kimi-k3
- Input · per 1M
- $3.00
- Output · per 1M
- $15.00
- Cached input · per 1M
- $0.30
- Context window
- 1,048,576 tokens
- Max output
- 1,048,576 tokens
- Knowledge cutoff
- Jan 2026
- Auto-router
- Not supported
Start building today
Route kimi-k3 — and every other model in the catalogue — through one endpoint, with failover built in.