Artificial Analysis said the model trailed Claude Fable 5 at 1574, beat GPT-5.6 Sol at 1501, and averaged $10.57 per task over 56.4 minutes using more tokens.
36d ago
verifying reliability
No specialized terms available for this topic.