Terminal-Bench 2.1 — leaderboard
State-of-the-art set of difficult terminal-based tasks.
Metric: Accuracy (%). Source: www.tbench.ai. Status: saturation imminent. 12 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude Fable 5 | 83.8 |
| 2 | GPT-5.5 | 83.2 |
| 3 | Grok 4.5 | 79.3 |
| 4 | Claude Opus 4.8 | 78.9 |
| 5 | GPT-5.6 Terra | 78.4 |
| 6 | Muse Spark 1.1 | 76.2 |
| 7 | GPT-5.6 Luna | 75.7 |
| 8 | Claude Sonnet 5 | 74.6 |
| 9 | Gemini 3 Pro | 73.9 |
| 10 | Claude Opus 4.7 | 68.9 |
| 11 | Gemini 3.1 Pro (Preview) | 65.8 |
| 12 | GLM-5.1 | 58.6 |
Interactive version: theaggregate.ai/benchmark?slug=terminal-bench-2-1 · How the rankings work · Data refreshed daily, snapshot 2026-07-22.