Claude Sonnet 5.5 Scores 70.6% on Terminal-Bench, Up from 10.3%
Anthropic's Claude Sonnet 5.5 is a faster, cheaper upgrade to Sonnet 5, scoring 70.6% on Terminal-Bench 4.0 versus 10.3% and costing up to 30% less per task. It beats Pokémon Red from screenshots and rivals Opus 5.5 on knowledge work, with cyber safeguards and preserved thinking.
In Epic's early testing, Claude Sonnet 5.5 cleared the same quality bar you'd expect from a higher-tier model, holding up on a system design audit and a data flow review. The new model managed tens of thousands of lines of code for gameplay system architecture, kept responses snappy, handled multi-hour tasks, and delivered with less prescriptive prompting.