Agents on Rails: leaderboard
Metric: Successful Runs (%). Source: rubyonrails.org. 17 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude Opus 5 | 92.1 |
| 2 | Claude Fable 5.1 (High) | 92.1 |
| 3 | Claude Fable 5 | 90.5 |
| 4 | Kimi K3 | 90.5 |
| 5 | GPT-5.6 Sol | 84.1 |
| 6 | Grok 4.6 (High) | 82.5 |
| 7 | GLM-5.3 Flash (Max) | 82.5 |
| 8 | GLM-5.3 (Max) | 79.4 |
| 9 | Claude Opus 4.8 (High) | 79.4 |
| 10 | GPT-5.6 Terra (Medium) | 77.8 |
| 11 | Muse Spark 1.2 | 76.2 |
| 12 | GPT-5.6 Luna | 73 |
| 13 | Gemini 3.7 Flash (Medium) | 71.4 |
| 14 | Claude Sonnet 5 (High) | 69.8 |
| 15 | GLM-5.2 | 66.7 |
Interactive version: theaggregate.ai/benchmark?slug=agents-on-rails · How It Works · Data refreshed daily, snapshot 2026-09-05.