Benchfolio

Leaderboard / Compare

Claude Opus 5.5 vs GPT-5.5

Claude Opus 5.5 wins both shared jobs.

Claude Opus 5.5GPT-5.5

Score by job

Scores run from 0 to 100 within each job. A gap under 3 points is too close to call. ~ marks an early read from too few benchmarks to rank. How scores work

At a glance

DetailClaude Opus 5.5GPT-5.5
MakerAnthropicOpenAI
ReleasedSep 22, 2026Apr 23, 2026
WeightsClosedClosed
Price per 1M tokens$4 in · $20 out$5 in · $30 out
Overall rankNot ranked yet#9

Compare other models

First model
Second model