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 navigating large repositories, using tools, debugging, and iterating against images, logs, tests, and runtime feedback. Its architecture uses KDA and Attention Residuals for computational efficiency. Note: reasoning effort currently supports only the max level (default); more levels are coming soon. See Moonshot AI's official docs here
All providers = every upstream serving this model supports it. Some providers = depends on which upstream handles the request. Default = the value sent when you leave the parameter unset.
| Parameter | Providers | Default |
|---|---|---|
| frequency_penalty | All providers | Not sent by default |
| include_reasoning | All providers | - |
| max_tokens | All providers | - |
| presence_penalty | All providers | Not sent by default |
| reasoning | All providers | - |
| reasoning_effort | All providers | - |
| repetition_penalty | Some providers | Not sent by default |
| response_format | All providers | - |
| stop | All providers | - |
| structured_outputs | All providers | - |
| temperature | Some providers | Not sent by default |
| tool_choice | All providers | - |
| tools | All providers | - |
| top_k | Some providers | Not sent by default |
| top_p | Some providers | Not sent by default |
