Model
GPT-5.1 Codex Max
OpenAI · closed weights
Agentic / Tool Use
Agentic / Tool Use| Benchmark | Score | Metric | Harness | Source | Confidence | Link |
|---|---|---|---|---|---|---|
| Terminal-Bench | 60.4 | accuracy | Codex CLI | Independent | medium | tbench.ai |
| Terminal-Bench | 57.8 | accuracy | Crux | Independent | medium | tbench.ai |
| Terminal-Bench | 53.5 | accuracy | Letta Code | Independent | medium | tbench.ai |
| Terminal-Bench | 36.9 | accuracy | Terminus 2 | Independent | medium | tbench.ai |
Reproduced by the community
Loading community runs…