ComposioBENCH
Try Bench

Claude Code

Anthropic#5 of 8 by coverage
Compare harnesses
Best success72%with GPT-6 Astra
Fastest run198sper task with GPT-6 Astra
Model support
Open
Graded runs
2 of 2 models
Latest run
2026-10-15

Success rate

% of tasks passed by model · higher is better
GPT-6 Astra
72.4%
DeepSeek V4 Pro
50.0%

Time per task

Seconds per task by model · lower is better
GPT-6 Astra
198s
DeepSeek V4 Pro
414s

Runs

  1. 01GPT-6 Astra
    72%
    Time / task198s
  2. 02DeepSeek V4 Pro
    50%
    Time / task414s