Anthropic · Benchmark update
Anthropic publishes Claude Sonnet 5.5 launch results
Anthropic's launch results for Sonnet 5.5 include Terminal-Bench 4.0, where it reports 70.6% against Sonnet 5's 10.3%.
- Occurrence
- Occurrence date basis
- Provider publication date
- Status recorded with this event
- effective
- Event checked
Original sources
- Introducing Claude Sonnet 5.5 (opens in a new tab) ↗
First-party source · Source checked
Sonnet 5.5 scores 70.6% on Terminal-Bench 4.0, an agentic coding evaluation, compared to Sonnet 5’s 10.3%.
Recording provenance
First recorded by StackReplay . This is a discovery date, not the occurrence date.
Event ID: anthropic-sonnet-5-5-results