TestsBenchmarks · Human preference

Arena Hard

LLM-judged hard chat prompts.

The best 1 · 1 models tested

Top 1 · best setting per model · 1 models, 2 results (0 independent)

Measured by independent testersIndependent Reported by the makerVendor-reported

Arena Hard leaderboard
				Thinking levelSetting
1	Mistral Small 4	Mistral AI	58.3%	HighReasoning (reasoning_effort=high)	Maker's own figureVendor-reportedMistral AI ↗	—	16 Mar 2026