Kilo Code 4.14.4 Release: Now with GPT 4.1 Integration
Supercharge your coding workflow with direct access to the latest OpenAI model, now built right into your favorite development environment.
We're excited to announce the release of Kilo Code 4.14.4, continuing our commitment to providing developers with the most powerful AI coding assistant experience. This update introduces direct support for OpenAI's cutting-edge GPT-4.1 model, further enhancing our "superset approach" that provides you with access to the best AI models through a consistent interface.
These are just a few highlights of the improvements in this release. Read on to check out all the great updates below.
GPT-4.1 Integration
OpenAI's latest GPT-4.1 model is now fully integrated into Kilo Code, bringing improved code generation, better context handling, and more accurate technical assistance directly to your development workflow. This update represents a significant leap forward in AI assistance for coding tasks.
With GPT-4.1, you'll experience:
Enhanced code completion with better understanding of your codebase
More accurate suggestions for complex programming problems
Improved context retention for longer coding sessions
Better technical documentation generation
Using the new model is straightforward:
Open Kilo Code
Click on the “settings” gear icon
Under “Provider Type” select GPT 4.1
Continue coding with the power of the latest AI technology
No need to manage separate API keys or switch between different tools - it's all seamlessly integrated within Kilo Code.
And we’d love to see what you build with Kilo Code and GPT-4.1 - share it in our Discord and let us know what you think!
Multi-Model Provider Support
Kilo Code continues to build on our foundation as a superset of AI coding capabilities. With this release, we now offer support for:
OpenAI GPT-4.1
Claude Sonnet 3.7
Gemini 2.5
This means you can easily switch between different AI providers without changing your workflow or managing multiple tools. You can even set up different configuration profiles to switch between models on the fly! We handle all the integration complexities so you can focus on writing great code (and vibes).
Bug Fixes & Usability Enhancements
Removed Deprecated Quasar Option
We've removed the Quasar Alpha option from the model selection as it's no longer available through OpenRouter - because it has been revealed to be an early test of GPT-4.1! This lined up with our (and many others) assumptions about this model.
This cleanup ensures you only see currently supported models in your interface.
Fixed Feedback Button Handler
We've resolved an issue with the feedback button functionality that was preventing some users from submitting their thoughts on Kilo Code. Your feedback is crucial to our improvement process, so please continue sharing your experiences with us.
Additional Improvements
Backend optimizations for improved performance when switching between models
Updated dependencies for enhanced security and stability
Streamlined model selection interface for a better user experience
For a full list of changes, you can always visit our GitHub Releases Page.