Benchmark · Knowledge

MultiChallenge

Updated 2026-02-19
Models tested
1
Top score
71.4
Gemini 3.1 Pro Preview
Median
71.4
min 71.4
Top-5 spread
σ 0.0
Settled

1 models tested · sorted by score

#ModelScore
1Google DeepMind logoGemini 3.1 Pro Preview71.4

Same category · related evaluations