GitHub Copilot Adds Configurable Suggestion Eagerness
GitHub Copilot in VS Code now lets developers set inline-suggestion eagerness to Low, Medium, or High, controlling how proactively it offers code. The control builds on Copilot’s unified inline-suggestions model, which combines ghost text, nearby edits, and farther-away edits in one system.
This is a smart UX correction: autocomplete should adapt to a developer’s rhythm instead of imposing one interruption policy. The unified model makes that personalization more meaningful because eagerness changes when Copilot speaks, not which disconnected model handles the request.
- –Low should favor uninterrupted typing; High suits developers who want more proactive edits, while Medium is the practical default.
- –One model can choose between ghost text, nearby rewrites, farther edits, cross-file changes, or no suggestion.
- –Online testing showed a 10.1% reduction in dismissals without statistically significant regressions in key metrics.
- –The real win is user agency: higher eagerness may improve discovery, but lower eagerness protects focus and reduces cognitive load.
DISCOVERED
1h ago
2026-10-26
PUBLISHED
2h ago
2026-10-26
RELEVANCE
AUTHOR
code