Release v4.111.1

[v4.111.1]

• #3337 85a4447 Thanks @mcowger! - Added MiniMax M2 to Synthetic Provider

• #3282 ed4399b Thanks @chrarnoldus! - Improved handling of tool calls in the API conversation history

• #3270 2b35053 Thanks @chrarnoldus! - Claude Haiku 4.5 now uses a simplified read file tool for reduced error rate

Release v4.111.0

[v4.111.0]

• #3256 f81b48b Thanks @markijbema! - Switched autocomplete to showing completions inline

Patch Changes

• #3261 bae048f Thanks @mcowger! - Improve native tool calling consistency

• #3281 2586e9b Thanks @hassoncs! - Fix broken sign-in links

• #3313 2e61e91 Thanks @chrarnoldus! - Running commands are no longer sometimes shown twice in the chat

Release v4.110.0

[v4.110.0]

• #3104 3008656 Thanks @markijbema! - Support multiline quoted strings within Auto-approved commands (especially useful for git commit)

Patch Changes

• #3249 ccee64c Thanks @chrarnoldus! - Remove credit status bar until we can retrieve the up to date balance from the proxy response.

• #3235 0108896 Thanks @chrarnoldus! - Fixed some "not a function" crashes

• #3226 e13a99c Thanks @chrarnoldus! - Reverted "Update base URL for Vertex Anthropic models to work around outdated library." because it is causing issues for some users.

• #2663 43140c9 Thanks @NaccOll! - Fix listCodeDefinitionNamesTool for annotated Java methods

• #3242 8604c83 Thanks @chrarnoldus! - Improved support for HTTP proxy environment variables

Release v4.109.2

[v4.109.2]

• #3216 3f34635 Thanks @markijbema! - Do not accept an autocomplete suggestion with shift-tab or ctrl-tab (only plain tab)

• #3214 b271af9 Thanks @mcowger! - Update Synthetic Provider to support GLM 4.6, and enable native tool calling

• #3199 14bbc5f Thanks @possible055! - Improve Chinese translation of autocomplete-related terms

Release v4.109.1

[v4.109.1]

• #3203 aeb8bf3 Thanks @chrarnoldus! - Fix Z.ai provider giving an Unauthorized error

Release v4.109.0

[v4.109.0]

• #3088 84a1fa3 Thanks @mcowger! - Update base URL for Vertex Anthropic models to work around outdated library.

• #3192 7015c23 Thanks @markijbema! - Fix bug: autocomplete no longer suggests strange XML

Patch Changes

• #3159 935bbae Thanks @kevinvandijk! - Include changes from Roo Code v3.28.18

  • Fix: Remove request content from UI messages to improve performance and reduce clutter (#5601 by @MuriloFP, #8594 by @multivac2x, #8690 by @hannesrudolph, PR by @mrubens)
  • Fix: Add userAgent to Bedrock client for version tracking (#8660 by @ajjuaire, PR by @app/roomote)
  • Feat: Z AI now uses only two coding endpoints for better performance (#8687 by @hannesrudolph)
  • Feat: Update image generation model selection for improved quality (thanks @chrarnoldus!)

• #3194 b566965 Thanks @markijbema! - Do not trigger autocomplete for external events, like git changes

• #3100 3e409b8 Thanks @markijbema! - Added Amazon Bedrock as a provider for autocomplete

• #3149 79c7d60 Thanks @catrielmuller! - Jetbrains - Update framework (Plugin now requires IntelliJ IDEA 2024.3 or later)

• #3195 93371d0 Thanks @chrarnoldus! - Fixed crash when browser tool is used with native tool calling enabled

Release v4.108.0

[v4.108.0]

• #2674 2836aed Thanks @mcowger! - add send message on enter setting with configurable behavior

• #3090 261889f Thanks @mcowger! - Allow the use of native function calling for OpenAI-compatible, LM Studio, Chutes, DeepInfra, xAI and Z.ai providers.

Patch Changes

• #3155 6242b03 Thanks @NikoDi2000! - Improved the Chinese translation of "run" from '命令' to '运行'

• #3120 ced4857 Thanks @mcowger! - The apply_diff tool was implemented for experimental JSON-style tool calling

Release v4.107.0

[v4.107.0]

• #2549 4327563 Thanks @mcowger! - Add Kilocode Credits Status Bar Item

Patch Changes

• #3082 d82e684 Thanks @chrarnoldus! - The "Enable browser tool" setting is now honored when using experimental JSON style tool calls.

• #3059 d71f1d6 Thanks @hassoncs! - Fix bug preventing the agent from editing files properly when git diff views are open

• #3105 b0c7475 Thanks @metju90! - Fix button styling on Let's Go CTA

• #3107 c58c4ac Thanks @kevinvandijk! - Include changes from Roo Code v3.28.16-v3.28.17

  • Fix: Update zh-TW run command title translation (thanks @PeterDaveHello!)
  • feat: Add Claude Sonnet 4.5 1M context window support for Claude Code (thanks @ColbySerpa!)

Release v4.106.0

[v4.106.0]

• #2833 0b8ef46 Thanks @mcowger! - (also thanks to @NaccOll for paving the way) - Preliminary support for native tool calling (a.k.a native function calling) was added.

  This feature is currently experimental and mostly intended for users interested in contributing to its development.
  It is so far only supported when using OpenRouter or Kilo Code providers. There are possible issues including, but not limited to:

  • Missing tools (e.g. apply_diff tool)
  • Tools calls not updating the UI until they are complete
  • Tools being used even though they are disabled (e.g. browser tool)
  • MCP servers not working
  • Errors specific to certain inference providers

  Native tool calling can be enabled in Providers Settings > Advanced Settings > Tool Call Style > JSON.
  It is enabled by default for Claude Haiku 4.5, because that model does not work at all otherwise.

• #3050 357d438 Thanks @markijbema! - CMD-I now invokes the agent so you can give it more complex prompts

Release v4.105.0

[v4.105.0]

• #3005 b87ae9c Thanks @kevinvandijk! - Improve the edit chat area to allow context and file drag and drop when editing messages. Align more with upstream edit functionality

Patch Changes

• #2983 93e8243 Thanks @jrf0110! - Adds project usage tracking for Teams and Enterprise customers. Organization members can view and filter usage by project. Project identifier is automatically inferred from .git/config. It can be overwritten by writing a .kilocode/config.json file with the following contents:

  {
  	"project": {
  		"id": "my-project-id"
  	}
  }

• #3057 69f5a18 Thanks @chrarnoldus! - Thanks Roo, support for Claude Haiku 4.5 to Anthropic, Bedrock and Vertex providers was added

• #3046 1bd934f Thanks @chrarnoldus! - A warning is now shown when the webview memory usage crosses 90% of the limit (gray screen territory)

• #2885 a34dab0 Thanks @shameez-struggles-to-commit! - Update VS Code Language Model API provider metadata to reflect current model limits:

  • Align context windows, prompt/input limits, and max output tokens with the latest provider data for matching models: gpt-3.5-turbo, gpt-4o-mini, gpt-4, gpt-4-0125-preview, gpt-4o, o3-mini, claude-3.5-sonnet, claude-sonnet-4, gemini-2.0-flash-001, gemini-2.5-pro, o4-mini-2025-04-16, gpt-4.1, gpt-5-mini, gpt-5.
  • Fixes an issue where a default 128k context was assumed for all models.
  • Notable: GPT-5 family now uses 264k context; o3-mini/o4-mini, Gemini, Claude, and 4o families have updated output and image support flags. GPT-5-mini max output explicitly set to 127,805.

  This ensures Kilo Code correctly enforces model token budgets with the VS Code LM integration.