Noometry

Math benchmark

FrontierMath Tier 4 leaderboard

As of October 2026, GPT-6.1 Sol has the highest published FrontierMath Tier 4 score on Noometry at 100%, out of 63 models with results.

Last verified

About FrontierMath Tier 4

The hardest FrontierMath tier: research-level problems written by professional mathematicians.

Category
Math
Introduced
2025
Format
Exact answer
Unit
Percent (random guessing ≈ 0%)
Official site
epoch.ai

Top 15 models

  1. GPT-6.1 Sol 100%
  2. GPT-6 Astra 97.6%
  3. Claude Opus 5.5 95%
  4. Claude Fable 5 90.2%
  5. GPT-6 Sol 90%
  6. Claude Fable 5.1 87.8%
  7. GPT-5.6 Sol 82.9%
  8. Claude Sonnet 5.5 80.5%
  9. GPT-5.5 Pro 78%
  10. AI Co-Mathematician 75.6%
  11. Claude Opus 5 73.2%
  12. GPT-5.5 72.5%
  13. GPT-5.6 Terra 70.7%
  14. GPT-5.6 Luna 61%
  15. GPT-5.4 Pro 58.5%

Sponsored placements are available on pages like this one. Advertise on Noometry

All results

#ModelProviderScoreSettingSourceDate
1GPT-6.1 Sol OpenAI100%maxEpoch AI2026-09-29
2GPT-6 Astra OpenAI97.6%highEpoch AI2026-08-30
3Claude Opus 5.5 Anthropic95%maxEpoch AI2026-09-22
4Claude Fable 5 Anthropic90.2%maxEpoch AI2026-06-09
5GPT-6 Sol OpenAI90%maxEpoch AI2026-09-22
6Claude Fable 5.1 Anthropic87.8%maxEpoch AI2026-09-01
7GPT-5.6 Sol OpenAI82.9%maxEpoch AI2026-07-09
8Claude Sonnet 5.5 Anthropic80.5%maxEpoch AI2026-09-29
9GPT-5.5 Pro OpenAI78%xhighEpoch AI2026-06-12
10AI Co-Mathematician Google75.6%Epoch AI2026-06-12
11Claude Opus 5 Anthropic73.2%maxEpoch AI2026-07-24
12GPT-5.5 OpenAI72.5%xhighEpoch AI2026-06-11
13GPT-5.6 Terra OpenAI70.7%maxEpoch AI2026-07-09
14GPT-5.6 Luna OpenAI61%maxEpoch AI2026-07-09
15GPT-5.4 Pro OpenAI58.5%xhighEpoch AI2026-06-13
16Claude Opus 4.8 Anthropic56.1%maxEpoch AI2026-06-10
17GPT-6 Luna OpenAI56.1%maxEpoch AI2026-09-22
18GPT-5.4 OpenAI49%xhighEpoch AI2026-06-11
19Claude Haiku 5.5 Anthropic46.3%maxEpoch AI2026-10-07
20Muse Spark 1.3 Meta46.3%maxEpoch AI2026-09-18
21Qwen3.8 Max Alibaba (Qwen)46.3%xhighEpoch AI2026-08-04
22GPT-5.2 Pro OpenAI46%xhighEpoch AI2026-06-13
23Kimi K3 Moonshot AI39%maxEpoch AI2026-07-17
24Gemini 3.7 Flash Google36.6%highEpoch AI2026-08-14
25Qwen3.7 Max Alibaba (Qwen)34.1%Epoch AI2026-06-13
26Claude Opus 4.7 Anthropic31.7%maxEpoch AI2026-06-10
27Grok 4.6 xAI31.7%xhighEpoch AI2026-08-14
28GPT-5.2 OpenAI31.7%xhighEpoch AI2026-06-11
29Claude Sonnet 5 Anthropic29.3%maxEpoch AI2026-06-30
30GLM-5.2 Z.ai (Zhipu)29.3%maxEpoch AI2026-06-19
31GLM-5.3 Z.ai (Zhipu)29.3%maxEpoch AI2026-08-25
32Claude Opus 4.6 Anthropic26.8%maxEpoch AI2026-06-11
33DeepSeek V4.1 Flash DeepSeek26.8%maxEpoch AI2026-10-08
34DeepSeek V4 Pro DeepSeek26.8%maxEpoch AI2026-08-19
35Gemini 3.1 Pro Preview Google26.8%Epoch AI2026-06-11
36Gemini 3.5 Flash Google26.8%highEpoch AI2026-06-10
37Kimi K2.6 Moonshot AI25.6%Epoch AI2026-06-10
38DeepSeek V4 Flash DeepSeek24.4%maxEpoch AI2026-08-02
39Grok 4.5 xAI24.4%highEpoch AI2026-07-09
40Gemini 3.6 Flash Google22%highEpoch AI2026-08-02
41Gemini 3.8 Flash Google22%highEpoch AI2026-09-02
42GPT-5 OpenAI22%highEpoch AI2026-06-11
43GPT-5 Pro OpenAI19.5%highEpoch AI2026-06-12
44Gemini 3 Flash Preview Google17.1%Epoch AI2026-06-11
45GLM-5.3-Flash Z.ai (Zhipu)17.1%maxEpoch AI2026-08-27
46Grok 4.20 (Non-Reasoning) xAI17.1%Epoch AI2026-07-13
47Grok 4.7 xAI17.1%xhighEpoch AI2026-09-22
48Inkling-Small Thinking Machines Lab17.1%xhighEpoch AI2026-08-14
49Grok 4.3 xAI14.6%highEpoch AI2026-06-17
50GPT-5.4 nano OpenAI12.2%highEpoch AI2026-06-12
51GPT-5 Mini OpenAI12.2%highEpoch AI2026-06-12
52Kimi K2.7 Code Moonshot AI12.2%Epoch AI2026-06-13
53GPT-5.4 mini OpenAI9.8%xhighEpoch AI2026-06-12
54Claude Opus 4.5 Anthropic4.9%32KEpoch AI2026-06-11
55Inkling Thinking Machines Lab4.9%xhighEpoch AI2026-08-06
56o4-mini OpenAI4.9%highEpoch AI2026-06-11
57Claude Opus 4.1 Anthropic2.4%32KEpoch AI2026-06-11
58Claude Sonnet 4.5 Anthropic2.4%32KEpoch AI2026-06-11
59GPT-5.5 Instant OpenAI2.4%Epoch AI2026-08-02
60GPT-5 Nano OpenAI2.4%highEpoch AI2026-06-12
61Gemini 2.5 Pro Google0%Epoch AI2026-06-11
62Gemini 3.5 Flash Lite Google0%highEpoch AI2026-08-02
63o3-mini OpenAI0%highEpoch AI2026-06-11

Compare the leaders

Other math benchmarks

Frequently asked questions

What does FrontierMath Tier 4 measure?

The hardest FrontierMath tier: research-level problems written by professional mathematicians.

Which model has the highest FrontierMath Tier 4 score?

As of October 2026, GPT-6.1 Sol has the highest published FrontierMath Tier 4 score on Noometry at 100%, out of 63 models with results.

What is the best open-weight model on FrontierMath Tier 4?

Kimi K3 has the highest FrontierMath Tier 4 accuracy among open-weight models at 39%, ranking 23 of 63 overall.