CursorBench 3.2.0

Earlier CursorBench release; not comparable to 4.0.

% solvedHigher is betterCounts toward:Weights: Coding ×0.6The test's websiteOfficial page ↗

The best 6 · 6 models tested

Top 6 · best setting per model · 6 models, 14 results (1 independent)

Measured by independent testersIndependent Reported by the makerVendor-reported

CursorBench 3.2.0 leaderboard
				Thinking levelSetting
1	Claude Fable 5.1	Anthropic	73.4%	Maxadaptive thinking, effort=max	Maker's own figureVendor-reportedAnthropic ↗	Cursor agent	1 Sep 2026
2	Claude Fable 5	Anthropic	70.5%	Maxadaptive thinking, effort=max	Maker's own figureVendor-reportedAnthropic ↗	Cursor agent	1 Sep 2026
3	Claude Opus 5	Anthropic	70.0%	Maxadaptive thinking, effort=max	Maker's own figureVendor-reportedAnthropic ↗	Cursor agent	1 Sep 2026
4	Grok 4.6	SpaceXAI (formerly xAI)	69.9%	High	Maker's own figureVendor-reportedSpaceXAI ↗	—	12 Aug 2026
5	GPT-5.6 Sol	OpenAI	67.2%	Maxreasoning effort=max	Independent testIndependentCursor (via Anthropic launch post) ↗	Cursor agent	1 Sep 2026
6	Grok 4.5	SpaceXAI (formerly xAI)	66.7%	High	Maker's own figureVendor-reportedSpaceXAI ↗	—	12 Aug 2026