ComposioBENCH
Try Bench

Codex

OpenAI#1 of 7 by coverage
Compare harnesses
Best success72%with GPT-6 Astra
Fastest run151sper task with GPT-6 Astra
Model support
Open
Graded runs
2 of 2 models
Latest run
2026-09-10

Success rate

% of tasks passed by model · higher is better
GPT-6 Astra
72.4%
DeepSeek V4 Pro
53.3%

Time per task

Seconds per task by model · lower is better
GPT-6 Astra
151s
DeepSeek V4 Pro
482s

Runs

  1. 01GPT-6 Astra
    72%
    Time / task151s
  2. 02DeepSeek V4 Pro
    53%
    Time / task482s