Ruby on Rails
So how did the four new models stack up? A quick look: • ox-alpha: This new stealth-mode model just dropped and passed 52 out of 63 tasks (82.5%), tied with Grok 4.6 for 5th place on the leaderboard. • @OpenAI Terra passed 49 of 63 tasks (77.8% accuracy) for just $0.20/run,
Ruby on Rails
Agents on Rails: lemans is now open source, and today four new models hit the benchmark: @openai’s Terra, @Alibaba_Qwen 3.8-27B, @AnthropicAI's Sonnet 5, and ox-alpha, a new model just released in stealth mode. As of August 24, 2026: - Fastest: @openAI Terra is now the fastest model, at a median of 3 minutes 2 seconds per task. - Most accurate: Still @claudeai Opus 5 by @AnthropicAI. Solved 92% of runs (58 of 63), and has the best API recall of all the models (35%). - Cheapest: @OpenAI GPT-5.6 Luna is still the cheapest from the models we can accurately price - just .012 cents per run. Read the latest benchmark report from @evilmartians to learn about lemans and see how the new models stacked up: https://rubyonrails.org/2026/8/24/agents-on-rails-lemans
