{"version":"0.68.0","versions":{"cli":"0.68.0","desktop":"0.22.0"},"date":"2026-03-04","dateFormatted":"March 4, 2026","url":"https://docs.factory.ai/changelog/release-notes#v0.68.0","jsonUrl":"https://docs.factory.ai/changelog/releases/0.68.0.json","summary":"Subagent inactivity timer, Usage Limits settings, and streaming stability fixes","surfaces":["cli"],"changes":[{"type":"feature","surface":"cli","title":"Subagent inactivity timer","description":"Subagents now automatically stop after 3 minutes of inactivity to prevent idle resource usage"},{"type":"feature","surface":"cli","title":"Usage Limits settings","description":"Configure and view Usage Limits directly within sessions and settings (Factory App)"},{"type":"fix","surface":"cli","title":"VS Code auto-installation","description":"Fixed embedded VS Code auto-installation in the Factory App"},{"type":"fix","surface":"cli","title":"Factory App stability","description":"Fixed daemon restart issues in the Factory App"},{"type":"fix","surface":"cli","title":"Session resume","description":"Fixed reliability issues when resuming previous sessions"},{"type":"fix","surface":"cli","title":"Streaming stability","description":"Fixed a crash that could occur during streaming retries"},{"type":"fix","surface":"cli","title":"BYOK error handling","description":"Improved error messages for payment and authentication errors with custom API keys"},{"type":"fix","surface":"cli","title":"Context management","description":"Improved handling when conversations exceed context limits"},{"type":"fix","surface":"cli","title":"Image handling for text-only models","description":"Images are now properly handled when using text-only models"},{"type":"fix","surface":"cli","title":"GLM-4.7 error handling","description":"Fixed error handling for GLM-4.7 model errors"},{"type":"fix","surface":"cli","title":"MCP tool compatibility","description":"Improved MCP tool schema sanitization for better compatibility across model providers"},{"type":"fix","surface":"cli","title":"Custom model streaming","description":"Fixed streaming issues with custom models"},{"type":"fix","surface":"cli","title":"Reasoning level display","description":"Fixed reasoning level not displaying correctly after using /model in spec mode"},{"type":"fix","surface":"cli","title":"Authentication reliability","description":"Reliability improvements to authentication mechanisms"}],"contentText":"New features\n\n- Subagent inactivity timer - Subagents now automatically stop after 3 minutes of inactivity to prevent idle resource usage\n- Usage Limits settings - Configure and view Usage Limits directly within sessions and settings (Factory App)\n\nBug fixes\n\n- VS Code auto-installation - Fixed embedded VS Code auto-installation in the Factory App\n- Factory App stability - Fixed daemon restart issues in the Factory App\n- Session resume - Fixed reliability issues when resuming previous sessions\n- Streaming stability - Fixed a crash that could occur during streaming retries\n- BYOK error handling - Improved error messages for payment and authentication errors with custom API keys\n- Context management - Improved handling when conversations exceed context limits\n- Image handling for text-only models - Images are now properly handled when using text-only models\n- GLM-4.7 error handling - Fixed error handling for GLM-4.7 model errors\n- MCP tool compatibility - Improved MCP tool schema sanitization for better compatibility across model providers\n- Custom model streaming - Fixed streaming issues with custom models\n- Reasoning level display - Fixed reasoning level not displaying correctly after using /model in spec mode\n- Authentication reliability - Reliability improvements to authentication mechanisms"}