From the VS Code blog
Latest
MAI-Code-1-Flash: early results from real developer workflows
MAI-Code-1-Flash is a lightweight coding model purpose built for the fast, iterative work developers do every day in GitHub. In this post, we share what the data is telling us about how it performs in real developer workflows.
Read post →
Recent posts
How Prompt Tuning Improved GPT-5.5 in VS Code
See how VS Code and OpenAI tested GPT-5.5 system prompt changes in a two-week experiment, cutting tool calls and tail-end token usage while speeding up edits.
Iterating faster with TypeScript 7
How the VS Code and TypeScript teams collaborated to adopt TypeScript 7 and speed up VS Code development
What 50,000 Runs of a 5-Line Eval Taught Us
How AI coding models calibrate effort, token cost, and tool use on even the simplest task, and what that means for model selection and cost.