Agents on Rails: Gemini 4 Argon enters top four but is slower and pricier
Gemini 4 Argon (high) ranks in the top four with a 60.6% success rate but is slower than OpenAI models and still costly even at a 50% discount, while Meta Muse Wonder reaches only 46.6%.
Agents on Rails update: Muse Wonder and Gemini 4 Argon have now been tested. - @Meta Muse Wonder (xhigh) had only a 46.6% success rate, landing in the bottom half of the results. - @GeminiApp Gemini 4 Argon (high) is now in the top 4 performers with a 60.6% success, but it is slower than OpenAI models. Both models landed on the expensive end of the charts, and it's worth noting that Gemini 4 Argon was even run at a 50% discounted rate (which ends in December). GPT-6.1 Sol is still the best bang for your buck. Check out the results at: https://rubyonrails.org/ai