Agents on Rails - Feature Tickets: leaderboard

Metric: Successful Runs (%). Source: rubyonrails.org. Saturation forecast: Around September 2028. 25 models tracked.

Top models

#ModelScore
1GPT-6 Astra (Max)53.33
2GPT-6 Astra (Medium)35
3Claude Opus 5.5 (Medium)33.33
4Claude Opus 5 (Max)31.67
5Claude Fable 5.1 (Max)31.67
6Grok 4.7 (High)31.67
7Claude Fable 5.1 (High)31.67
8GPT-5.6 Sol (Max)28.33
9Gemini 3.8 Flash (Medium)28.33
10GPT-5.6 Luna (Max)26.67
11Claude Opus 5 (High)25
12Gemini 3.8 Flash (High)23.33
13GPT-6 Sol (Medium)21.67
14Muse Spark 1.3 (Max)20
15GPT-6 Luna (Max)18.33

Interactive version: theaggregate.ai/benchmark?slug=agents-on-rails-feature-tickets · How It Works · Data refreshed daily, snapshot 2026-10-09.