Does anyone know if these benchmark values can still be compared? They differences seem rather small or what am I missing? Thanks! https://artificialanalysis.ai/models/gpt-6-1-sol?models=gpt-...
[dead]
[dead]