Model Configuration | Kimi Code Docs
Skip to contentMenuReturn to top
Model Configuration <br>This page covers the models Kimi Code provides and how to switch between them in each client.<br>Model Overview <br>Kimi Code currently offers two models—Kimi K3 and Kimi K2.7 Code—across three model IDs, selectable by model ID in clients or third-party tools. Model specs:<br>Model IDk3kimi-for-codingkimi-for-coding-highspeedModelKimi K3Kimi K2.7 CodeKimi K2.7 CodeDescriptionKimi's most capable flagship model to date; especially strong at coding, games/3D, and knowledge tasksA mature, reliable Coding model—follows instructions reliably over long context and completes coding tasks with a high success rateThe high-speed version of K2.7 Code, with the same coding ability as StandardSpeedRegularRegularHighSpeed (6× speed, 3× quota usage)Context windowUp to 1M256k256kReasoningthinking.effort:max / reasoning_effort:max (low and high supported later)Thinking:ONThinking:ONAvailabilityAndante: not supported<br>Moderato: 256k context<br>Allegretto and above: up to 1M contextAll membersAllegretto plan or aboveNeed a higher membership plan?<br>Different membership plans unlock different models, context windows, and speeds. Upgrade your plan →
Why did usage go up after the new model launched?After switching models, the context cache built earlier no longer hits on the new model, so that context has to be re-prefilled. Usage therefore looks higher right after switching. Recommended action:<br>Start a new session when using the new model : this gives better results and lower consumption.<br>Why do I still get a 401 with the correct model ID?When the requested capability exceeds your plan's entitlements, the server returns 401. Three common cases:<br>No K3 access : your plan is below Moderato and can't call k3 — upgrade to Moderato or above.<br>No 1M access : on a Moderato plan, k3 supports up to 256K context; up to 1M context is available on Allegretto and higher tiers.<br>No HighSpeed access : some plans don't include HighSpeed — upgrade to Allegretto or a higher tier to call kimi-for-coding-highspeed.<br>For the full error text and how to handle it, see the Error Reference.<br>Why isn't HighSpeed noticeably faster?Two common reasons:<br>Mistyped model ID : the HighSpeed ID must be kimi-for-coding-highspeed; a wrong value silently falls back to the standard kimi-for-coding — no error, no speedup.<br>Tools and scripts dominate : HighSpeed only speeds up model output . Tool calls (reading/writing files, running commands, etc.) and script execution are unaffected, so when they take up most of a turn the overall speedup feels small.<br>How to reduce the overhead of switching reasoning effort?Switching reasoning effort invalidates the context cache you've built up, so context that would have hit the cache must be re-prefilled. To avoid triggering re-prefill too often:<br>Pick an effort that fits the task and keep it consistent within a session;<br>When you genuinely need a different effort, start a new session rather than switching back and forth in a long session.<br>How to Switch Models <br>Use K3 in a New Session<br>Switching models invalidates the existing cache. Start a new session to avoid extra token costs and get a better experience.
Ways to switch to the target model:<br>Official Kimi Code CLI : type /model in a session to switch between models—no config changes needed.<br>Kimi Code for VS Code : pick the target model from the dropdown menu in the input bar; if it isn't listed yet, restart VS Code or reinstall the extension.<br>Third-party tools : set the tool's Model ID to the target model. For where to find it in each tool, see Using in Other Coding Agents.<br>Before using K3 in third-party tools<br>K3's setup differs slightly from K2.7 Code. Before using it, check the two points below:<br>Context window : some tools default to a context window smaller than the max 1M—manually set the context-window field to 1048576 to use K3's full up-to-1M context.<br>Reasoning effort : K3 currently supports max only; low and high will be supported later. The effort a tool sends is mapped as below:<br># default<br>null / undefined → max<br>any other unknown → HTTP 400 error
# → max<br>ultra / max / xhigh → max
# → high<br>high / medium → high
# → low<br>low / minimum / light → low
# → thinking disabled<br>none → thinking.type disabled