ComposioBENCH
Try Bench

Cline

Cline#1 of 8 by coverage
Compare harnesses
Best success79%with GPT-6 Astra
Fastest run193sper task with GPT-6 Astra
Model support
Open
Graded runs
2 of 2 models
Latest run
2026-10-16

Success rate

% of tasks passed by model · higher is better
GPT-6 Astra
79.3%
DeepSeek V4 Pro
56.7%

Time per task

Seconds per task by model · lower is better
GPT-6 Astra
193s
DeepSeek V4 Pro
327s

Runs

  1. 01GPT-6 Astra
    79%
    Time / task193s
  2. 02DeepSeek V4 Pro
    57%
    Time / task327s