Skip to content

Sign in with GitHub

Terminal-Bench 2.0 on GPT-5.6 Sol

89 tasks, 1 attempt, open network. Updated 24 Oct 2026.

Build a run for this board

Pass rate by harness, higher is better

HarnessPass rate

Codex0.163.062.9%95% interval 52.8–73.0

All harnesses

Reading this table

Every harness on this board, in its order
Harness	Runs	Last published
Codex0.163.0	13 infra2 refused	24 Oct 2026run #21

Every result is self-reported
Data as of 13 Nov 2026
Methodology
API
Report a submission