BIG-Bench Extra Hard

Harder successor to BIG-Bench Hard reasoning tasks.

The best 2 · 2 models tested

Top 2 · best setting per model · 2 models, 2 results (0 independent)

Measured by independent testersIndependent Reported by the makerVendor-reported

BIG-Bench Extra Hard leaderboard
				Thinking levelSetting
1	Gemma 4 31B IT	Google (Gemini / DeepMind)	74.4%	DefaultThinking	Maker's own figureVendor-reportedGoogle ↗	—	2 Apr 2026
2	Gemma 4 26B A4B IT	Google (Gemini / DeepMind)	64.8%	DefaultThinking	Maker's own figureVendor-reportedGoogle ↗	—	2 Apr 2026