Claude Sonnet 5.5
How-To How to actually use this
What changed: Anthropic released Claude Sonnet 5.5, a faster and cheaper drop-in replacement for Sonnet 5 with improved benchmark performance.
How to use it:
- Select "Claude Sonnet 5.5" in the model dropdown on the Anthropic Console, API, or claude.ai chat interface.
- Send prompts exactly as you did for Sonnet 5; no code or prompt changes are required.
- Monitor usage dashboards to verify reduced latency and lower token costs for your workloads.
- If using "extended thinking" mode, set a max thinking token budget well below 128,000 to avoid the known runaway reasoning bug.
Good for: developers and teams currently using Sonnet 5 for coding, analysis, or high-volume chat.
Claude Sonnet 5.5 New Sonnet model from Anthropic today. They say it "runs 30%+ faster, and costs up to 30% less for most work" - it's priced the same as Sonnet 5 but appears to beat it on every benchmark, and should be cheaper to run as well. Here are some pelicans riding bicycles . Sonnet 5.5 suffered from the same bug as Opus 5.5 : the "max" thinking effort pelican thought for 128,000 tokens…
Read original article on Simon Willison's Weblog →More in Model Releases
