GPT-5-Codex

By OpenAI, in the Codex family.

Category scores

Out of 100, from independent and third-party results only, by method category-2026-09-29.1.

Not enough independent results yet

Only independent and third-party results rank a model; self-reported and unattributed numbers never do.

Every result we hold for this model, with who ran it and where it was published.

Coding & software engineering results
Benchmark	Metric	Value	Provenance	Setting	Evaluated	Source
Terminal-Bench 2.0	Accuracy (mean over trials)	44.3%	Third party	—	—	https://www.tbench.ai/leaderboard/terminal-bench/2.0 via Epoch AI
Terminal-Bench 2.0	Accuracy (mean over trials)	44.3%	Third party	—	—	Terminal-Bench v2 Leaderboard via Epoch AI
Terminal-Bench 2.0	Accuracy (mean over trials)	43.4%	Third party	—	—	https://www.tbench.ai/leaderboard/terminal-bench/2.0 via Epoch AI
Terminal-Bench 2.0	Accuracy (mean over trials)	43.4%	Third party	—	—	Terminal-Bench v2 Leaderboard via Epoch AI
Terminal-Bench 2.0	Accuracy (mean over trials)	41.3%	Third party	—	—	https://www.tbench.ai/leaderboard/terminal-bench/2.0 via Epoch AI
Terminal-Bench 2.0	Accuracy (mean over trials)	41.3%	Third party	—	—	Terminal-Bench v2 Leaderboard via Epoch AI
WeirdML (v2)	Average accuracy	54.5%	Third party	—	—	WeirdML Leaderboard via Epoch AI
WeirdML (v2)	Average accuracy	54.5%	Third party	high	—	WeirdML Leaderboard via Epoch AI