benchgap
Calibration

FrontierMath v2 (Tiers 1-3) → HMMT Nov 2025

HMMT Nov 2025 is estimated from FrontierMath v2 (Tiers 1-3) with a Michaelis–Menten curve fitted on 6 models measured on both: y = 0.9827·x / (0.02058 + x), R² = 0.69, cross-validated error 1.5 pp. It is used for 63 estimates.

Estimated modelFrontierMath v2 (Tiers 1-3)HMMT Nov 2025Source
Claude Fable 5.295.1%96.2%estimated ± 1.5 pp, low confidence
GPT-6.1 Sol93.7%96.2%estimated ± 1.5 pp, low confidence
GPT-6 Astra93.7%96.2%estimated ± 1.5 pp, low confidence
GPT-6.2 Astra92.6%96.1%estimated ± 1.5 pp, low confidence
Claude Opus 5.591.2%96.1%estimated ± 1.5 pp, low confidence
GPT-6.1 Terra90.9%96.1%estimated ± 1.5 pp, low confidence
Gemini 4 Carbon90.5%96.1%estimated ± 1.5 pp, low confidence
Claude Fable 5.190.2%96.1%estimated ± 1.5 pp, low confidence
GPT-6 Sol89.8%96.1%estimated ± 1.5 pp, low confidence
GPT-5.6 Sol89.0%96.1%estimated ± 1.5 pp, low confidence
Claude Sonnet 5.588.8%96.0%estimated ± 1.5 pp, low confidence
GPT-5.5 Pro87.7%96.0%estimated ± 1.5 pp, low confidence
Claude Fable 587.0%96.0%estimated ± 1.5 pp, low confidence
Gemini 4 Argon86.0%96.0%estimated ± 1.5 pp, low confidence
Claude Opus 585.6%96.0%estimated ± 1.5 pp, low confidence
GPT-5.585.3%96.0%estimated ± 1.5 pp, low confidence
GPT-5.6 Terra84.9%95.9%estimated ± 1.5 pp, low confidence
GPT-5.4 Pro82.5%95.9%estimated ± 1.5 pp, low confidence
Claude Opus 4.880.0%95.8%estimated ± 1.5 pp, low confidence
GPT-6 Luna78.9%95.8%estimated ± 1.5 pp, low confidence
GPT-5.6 Luna78.6%95.8%estimated ± 1.5 pp, low confidence
GPT-5.478.6%95.8%estimated ± 1.5 pp, low confidence
Claude Haiku 5.575.1%95.7%estimated ± 1.5 pp, low confidence
Qwen3.8 Max74.7%95.6%estimated ± 1.5 pp, low confidence
Muse Spark 1.374.0%95.6%estimated ± 1.5 pp, low confidence
Gemini 3.7 Flash71.6%95.5%estimated ± 1.5 pp, low confidence
Claude Opus 4.770.2%95.5%estimated ± 1.5 pp, low confidence
GLM-5.368.8%95.4%estimated ± 1.5 pp, low confidence
Gemini 3.8 Flash68.4%95.4%estimated ± 1.5 pp, low confidence
GPT-5.267.4%95.4%estimated ± 1.5 pp, low confidence
DeepSeek V4.1 Flash 101767.4%95.4%estimated ± 1.5 pp, low confidence
Claude Opus 4.666.0%95.3%estimated ± 1.5 pp, low confidence
Grok 4.666.0%95.3%estimated ± 1.5 pp, low confidence
Claude Sonnet 565.6%95.3%estimated ± 1.5 pp, low confidence
DeepSeek V4 Pro64.6%95.2%estimated ± 1.5 pp, low confidence
Qwen3.7 Max64.6%95.2%estimated ± 1.5 pp, low confidence
Gemini 3.5 Flash62.8%95.2%estimated ± 1.5 pp, low confidence
Gemini 3.1 Pro59.6%95.0%estimated ± 1.5 pp, low confidence
Gemini 3.6 Flash58.9%95.0%estimated ± 1.5 pp, medium confidence
Grok 4.557.2%94.9%estimated ± 1.5 pp, medium confidence
Kimi K2.657.2%94.9%estimated ± 1.5 pp, medium confidence
GLM-5.3-Flash55.8%94.8%estimated ± 1.5 pp, medium confidence
Kimi K2.7 Code54.0%94.7%estimated ± 1.5 pp, medium confidence
Grok 4.753.0%94.6%estimated ± 1.5 pp, medium confidence
Gemini 3 Flash51.2%94.5%estimated ± 1.5 pp, medium confidence
GPT-5.4 mini51.2%94.5%estimated ± 1.5 pp, medium confidence
Inkling-Small46.3%94.1%estimated ± 1.5 pp, medium confidence
GPT-5.4 nano44.9%94.0%estimated ± 1.5 pp, medium confidence
Grok 4.342.8%93.8%estimated ± 1.5 pp, medium confidence
o4-mini36.1%93.0%estimated ± 1.5 pp, medium confidence
Qwen3.7 Plus34.4%92.7%estimated ± 1.5 pp, medium confidence
Inkling33.3%92.6%estimated ± 1.5 pp, medium confidence
o333.3%92.6%estimated ± 1.5 pp, medium confidence
Gemini 3.1 Flash-Lite27.7%91.5%estimated ± 1.5 pp, medium confidence
Gemini 3.5 Flash-Lite26.0%91.1%estimated ± 1.5 pp, medium confidence
Gemini 2.5 Pro24.6%90.7%estimated ± 1.5 pp, medium confidence
Claude Sonnet 4.523.9%90.5%estimated ± 1.5 pp, medium confidence
Qwen3.7 Flash19.3%88.8%estimated ± 1.5 pp, low confidence
Qwen3.5 Flash18.2%88.3%estimated ± 1.5 pp, low confidence
o114.7%86.2%estimated ± 1.5 pp, low confidence
GPT-4.1 mini6.7%75.1%estimated ± 1.5 pp, low confidence
GPT-4.16.0%73.1%estimated ± 1.5 pp, low confidence
GPT-4o0.4%14.3%estimated ± 1.5 pp, low confidence

Fitted on

ModelFrontierMath v2 (Tiers 1-3)HMMT Nov 2025
Claude Opus 4.534.4%93.3%
GLM-5.136.8%94.0%
GLM-5.259.2%94.4%
Qwen3.6-27B35.1%90.7%
Qwen3.6-35B-A3B20.4%89.1%
Qwen3.6 Plus38.2%94.6%