TestsBenchmarks · Math

FrontierMath Tier 1-3

Epoch AI research-level math problems, tiers 1-3.

% solvedHigher is betterThe test's websiteOfficial page ↗

The best 5 · 5 models tested

Top 5 · best setting per model · 5 models, 6 results (0 independent)

Measured by independent testersIndependent Reported by the makerVendor-reported

FrontierMath Tier 1-3 leaderboard
				Thinking levelSetting
1	GPT-5.6 Sol	OpenAI	89.0%	DefaultGPT-5.6 launch table (effort not stated; text cites max for headline results)	Maker's own figureVendor-reportedOpenAI ↗	—	9 Jul 2026
2	GPT-5.5	OpenAI	85.3%	DefaultGPT-5.6 launch table (effort not stated; text cites max for headline results)	Maker's own figureVendor-reportedOpenAI ↗	—	9 Jul 2026
3	GPT-5.6 Terra	OpenAI	84.9%	DefaultGPT-5.6 launch table (effort not stated; text cites max for headline results)	Maker's own figureVendor-reportedOpenAI ↗	—	9 Jul 2026
4	GPT-5.6 Luna	OpenAI	78.6%	DefaultGPT-5.6 launch table (effort not stated; text cites max for headline results)	Maker's own figureVendor-reportedOpenAI ↗	—	9 Jul 2026
5	GPT-5.5 Pro	OpenAI	52.4%	Extra highGPT-5.5 Pro, reasoning effort=xhigh	Maker's own figureVendor-reportedOpenAI ↗	—	23 Apr 2026