Rankings
Each ranking profile weights the same evaluation dimensions differently. Scores factor in confidence and evidence volume — see the methodology page for how they’re calculated.
General AI Reasoning
A balanced profile weighting reasoning, knowledge and instruction-following most heavily.
Coding
Weighted toward coding capability and tool use, for developer-focused use cases.
Reasoning
Weighted toward multi-step reasoning and knowledge depth.
Value / Quality
Balances overall capability against cost efficiency and speed.