Kimi K3
Moonshot AI's model with a very long context for code and agents
free access available
- Context
- 256,000 tokens
- Type
- Text, with reasoning
- Input
- Text, images
- Developer
- Moonshot AI
- Strengths
- Long context, code
- Access
- ChatPlus PRO
Kimi K3 is the flagship of the Chinese startup Moonshot AI, known for its very long context. It handles large repositories, big documents and agentic scenarios well, holding context across the whole span.
Kimi was built around long context from the start, and K3 pushes that idea to the limit. Large repositories, big documentation, long conversations — it holds all of it at once.
That helps on agentic tasks: when work spans many steps, K3 doesn't forget what came at the beginning and doesn't drift off halfway. It also understands pictures, so diagrams and screenshots go to work too.
In essence it's a model for anyone who kept running out of context window with other models.
What it's good at
Long context
Holds very large documents and repositories in one pass.
Programming
Works confidently with code and large codebases.
Agentic tasks
Long multi-step scenarios without losing context.
Multimodality
Understands images alongside text.
Kimi K3 and every other model — one plan, one window.
Where it's strongest
Kimi next to the others
How Kimi K3 and the flagships of other labs look side by side. They're all in one ChatPlus subscription.
| Claude | DeepSeek | GPT | Gemini | Grok | GLM | Kimi | Qwen | Muse | GLM 5.3 Flash | |
|---|---|---|---|---|---|---|---|---|---|---|
| Developer | Anthropic | DeepSeek | OpenAI | xAI | Z.ai | Moonshot | Alibaba | Meta | Z.ai | |
| Model | Sonnet 4.6 | DeepSeek V4.1 Reasoner | GPT-6 Astra | Gemini 3 Pro | Grok 4 | GLM 5 | Kimi K3 | Qwen 3.8 Max | Muse Spark 1.3 | GLM 5.3 Flash |
| Context | 200K | 128K | 1M | 1M | 256K | 1M | 256K | 256K | 1M | 1M |
| Understands images | — | — | ||||||||
| Live web data | — | — | — | — | — | — | — | |||
| Open source | — | — | — | — | — | |||||
| Key strength | Text and code | Reasoning | Computer use | Multimodality | Timeliness | Code and agents | Context | Complex tasks | Agents and code | Code and agents |
| On ChatPlus |