Computer Anthology Terminal Tasks (Terminus-2) — leaderboard
Metric: pass@1 (%). Source: vetto.ai. 18 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude Opus 5 | 61.8 |
| 2 | GPT-5.6 Sol | 51.6 |
| 3 | Claude Fable 5 | 50.8 |
| 4 | Kimi K3 | 40 |
| 5 | GPT-5.5 | 38.6 |
| 6 | GLM-5.2 | 31.4 |
| 7 | Claude Opus 4.8 | 29.8 |
| 8 | Gemini 3.5 Flash | 22.8 |
| 9 | Hy3 | 18 |
| 10 | Gemini 3.1 Pro (Preview) | 17.6 |
| 11 | Kimi K2.7 Code | 15.4 |
| 12 | Claude Opus 4.6 | 14.8 |
| 13 | GPT-5.4 | 12 |
| 14 | DeepSeek V4 Pro | 10.4 |
| 15 | Qwen 3.7 Max | 9.8 |
Interactive version: theaggregate.ai/benchmark?slug=computer-anthology-terminal-tasks-terminus-2 · How It Works · Data refreshed daily, snapshot 2026-08-06.