Bolt: Claude 4.5 Model Selection
BoltBolt introduced in-editor model switching across the Claude 4.5 family — Haiku 4.5 (fast), Sonnet 4.5 (balanced default), and Opus 4.5 (most capable). Users can switch models for different task types, optimizing between speed, capability, and token efficiency.
Sources & Mentions
4 external resources covering this update
Claude 4.5 Model Selection Now Available in Bolt
Bolt expanded its AI model offering with in-editor model switching across the full Claude 4.5 family.
The Three Models
Claude Haiku 4.5 is the fastest and most token-efficient option, suited for rapid prototyping and quick edits. Claude Sonnet 4.5 is the default and balanced all-rounder for most development work. Claude Opus 4.5 is the most capable, delivering stronger performance on complex, multi-step problems.
How to Use Model Switching
Users can switch between models at any point. Bolt recommends changing models at thread boundaries — starting a new chat thread when switching models for a different task type.
Impact on Workflows
A session might start with Haiku for scaffolding, transition to Sonnet for feature development, and reach for Opus when tackling the most demanding parts of the codebase. This tiered approach lets teams optimize token consumption while preserving quality.