terminal-bench@2.0
terminal-bench@2.0 is run by Terminal-Bench. It has ranked 142 models, of which the catalogue holds 6, scoring from 69.4 to 84.7 on Accuracy.
measured by Terminal-Bench · the board itself
| Place | Model | Metric | Score |
|---|---|---|---|
| 1stof 142 | GPT-5.5 | Accuracy | 84.7 |
| 6thof 142 | Claude Opus 4.7 | Accuracy | 80.2 |
| 7thof 142 | Gemini 3.1 Pro Preview | Accuracy | 80.2 |
| 9thof 142 | GPT-5.3 Codex | Accuracy | 78.4 |
| 11thof 142 | Claude Opus 4.6 | Accuracy | 76.4 |
| 24thof 142 | gemini-3-pro-preview | Accuracy | 69.4 |
Read from the board on 2026-08-25