Cody for VS Code v1.18: Now with GPT-4o support

Cody for VS Code v1.18 is now available. This update adds GPT-4o support for Cody Pro users, the super-fast new model from OpenAI. We’ve also added a new way for you to add files to Cody chat context directly from your file tree.

GPT-4o support for Cody Pro

OpenAI just unveiled its latest flagship model, GPT-4o. It’s two times faster than GPT-4 Turbo while beating its general reasoning benchmark scores.

From our (very early) testing, GPT-4o is a super fast and powerful model to use with Cody, so we’ve rolled it out as an option for Cody Pro users. You can now select it from the model dropdown for chat + commands.

We’ll roll out GPT-4o to Cody Enterprise users in a future release. You can also try out GPT-4o in s0.dev.

The model selection window gets a refresh

We’ve given the model selection dropdown a new coat of paint. Models are now organized for their strengths—speed versus accuracy—so you can make more informed decisions as you try them out for different use cases.

Add files as context directly from the file tree

Like human developers, Cody is more capable when using context relevant to a task. Choosing context for Cody should be as easy as possible to make sure it always performs to its max potential.

You can now add files as context for Cody directly from the file tree. Just right-click a file → Add File to Cody Chat, and the file path will populate as an @-mentionin the chat window.

If you don’t already have a Cody chat window open, this option will appear as right click → New Chat with File.