Reaction
GPT-6.1-sol claimed to be 3x as cost-efficient as Opus 5.5 in internal tests
A user says Gemini 4 Argon matched Opus 5.5's cost efficiency, assuming introductory pricing, but ran twice as slowly.
TLDR
A user says internal cfo.ai evaluations put Gemini 4 Argon on par with Opus 5.5 for cost efficiency, assuming introductory pricing, but found it twice as slow. The user says GPT-6.1-sol was 3x as cost-efficient as either model, though slower than Opus 5.5 in that run. In an earlier evaluation, the user reported GPT-6.1-sol was 2.5x as efficient as Sonnet 5.5.
Combined views
6.9K
1 Source, first seen ago
