Test Run #7 Analysis

Comparing model performance for the GPQA 2026 benchmark.

Global Filters

Languages

Models

Tags

Overall Avg. Score

0.333

Best Model

Claude 4 Sonnet

Highest Model Score

0.424

Model Scores per Language