# FrontierMath

> Unpublished research-level maths problems with answers a computer can check.

Source: https://www.superintelligence.com.au/benchmarks/frontiermath/
Data updated 2026-10-07. Benchmark data from Epoch AI (CC BY) and official leaderboards.

Best score today: 94% (+20 pts in 12 months).
Human comparison: No single human baseline exists. The problems were written by research mathematicians, and many take a specialist hours or days. No group of people has sat the whole test.

## What it tests

Hundreds of unpublished problems written by research mathematicians, with answers that can be checked automatically. This site tracks tiers 1 to 3; a harder fourth tier is scored separately.

## Why it matters

The problems are unpublished, so progress signals mathematical reasoning rather than memory.

## Caveats

The problems are not public, but OpenAI commissioned them and has access to most of the problems and their solutions. Epoch AI, which runs the test, keeps a set back from OpenAI to check results against. Scores here are for the current version of the set, re-run on older models.

## Best score by quarter

| Quarter | Best score | Record held by |
| --- | --- | --- |
| Q1 2024 | 0% | GPT-3.5 Turbo (Jan 2024) |
| Q2 2024 | 1% | gpt-4-turbo-2024-04-09 |
| Q3 2024 | 1% | gpt-4-turbo-2024-04-09 |
| Q4 2024 | 15% | o1 (high) |
| Q1 2025 | 19% | o3-mini (high) |
| Q2 2025 | 36% | o4-mini (high) |
| Q3 2025 | 55% | GPT-5 (high) |
| Q4 2025 | 74% | GPT-5.2 Pro |
| Q1 2026 | 83% | GPT-5.4 Pro (xhigh) |
| Q2 2026 | 88% | GPT-5.5 Pro (xhigh) |
| Q3 2026 | 94% | GPT-6.1 Sol (max) |
| Q4 2026 | 94% | GPT-6.1 Sol (max) |

## Top scores right now

| Rank | Model | Developer | Score | Released |
| --- | --- | --- | --- | --- |
| 1 | GPT-6 Astra (max) | OpenAI | 94% | 2026-09-03 |
| 2 | GPT-6.1 Sol (max) | OpenAI | 94% | 2026-09-29 |
| 3 | Claude Opus 5.5 (max) | Anthropic | 91% | 2026-09-22 |
| 4 | Claude Fable 5.1 (max) | Anthropic | 90% | 2026-09-01 |
| 5 | GPT-6 Sol (max) | OpenAI | 90% | 2026-09-22 |
| 6 | GPT-5.6 Sol (max) | OpenAI | 89% | 2026-07-09 |
| 7 | Claude Sonnet 5.5 (max) | Anthropic | 89% | 2026-09-28 |
| 8 | GPT-5.5 Pro (xhigh) | OpenAI | 88% | 2026-04-23 |
| 9 | Claude Fable 5 (max) | Anthropic | 87% | 2026-06-09 |
| 10 | GPT-5.6 Terra (max) | OpenAI | 86% | 2026-07-09 |
| 11 | Claude Opus 5 (max) | Anthropic | 86% | 2026-07-24 |
| 12 | GPT-5.5 (xhigh) | OpenAI | 85% | 2026-04-23 |
| 13 | GPT-5.4 Pro (xhigh) | OpenAI | 83% | 2026-03-05 |
| 14 | GPT-5.6 Luna (max) | OpenAI | 82% | 2026-07-09 |
| 15 | Claude Opus 4.8 | Anthropic | 80% | 2026-05-28 |
| 16 | GPT-6 Luna (max) | OpenAI | 79% | 2026-09-22 |
| 17 | GPT-5.4 (xhigh) | OpenAI | 79% | 2026-03-05 |
| 18 | Qwen3.8 Max (xhigh) | Alibaba | 75% | 2026-08-02 |
| 19 | Muse Spark 1.3 (xhigh) | Meta AI | 74% | 2026-09-02 |
| 20 | Muse Spark 1.3 (max) | Meta AI | 74% | 2026-09-02 |
| 21 | GPT-5.2 Pro | OpenAI | 74% | 2025-12-11 |
| 22 | Kimi K3 (max) | Moonshot | 72% | 2026-07-16 |
| 23 | Gemini 3.7 Flash (high) | Google DeepMind | 72% | 2026-08-13 |
| 24 | Claude Opus 4.7 (max) | Anthropic | 70% | 2026-04-16 |
| 25 | GLM-5.3 (max) | Z.ai (Zhipu AI) | 69% | 2026-08-14 |
| 26 | Gemini 3.8 Flash (high) | Google DeepMind | 68% | 2026-09-02 |
| 27 | GPT-5.2 (xhigh) | OpenAI | 67% | 2025-12-11 |
| 28 | Claude Opus 4.6 (max) | Anthropic | 66% | 2026-02-05 |
| 29 | Grok 4.6 (xhigh) | xAI | 66% | 2026-08-12 |
| 30 | Claude Sonnet 5 (max) | Anthropic | 66% | 2026-06-30 |
| 31 | Qwen3.8 Max (0902) (xhigh) | Alibaba | 66% | 2026-09-01 |
| 32 | Qwen3.7 Max | Alibaba | 65% | 2026-05-19 |
| 33 | DeepSeek V4 Pro 0813 (max) | DeepSeek | 65% | 2026-08-13 |
| 34 | Gemini 3.5 Flash | Google DeepMind | 63% | 2026-05-19 |
| 35 | Gemini 3.1 Pro Preview | Google DeepMind | 60% | 2026-02-19 |
| 36 | GLM-5.2 | Z.ai (Zhipu AI) | 59% | 2026-06-16 |
| 37 | Gemini 3.6 Flash | Google DeepMind | 59% | 2026-07-21 |
| 38 | DeepSeek V4 Flash 0731 (max) | DeepSeek | 58% | 2026-07-31 |
| 39 | Kimi K2.6 | Moonshot | 57% | 2026-04-20 |
| 40 | Grok 4.5 (high) | xAI | 57% | 2026-07-08 |
| 41 | GPT-5 Pro | OpenAI | 56% | 2025-10-07 |
| 42 | GLM-5.3-Flash (max) | Z.ai (Zhipu AI) | 56% | 2026-08-20 |
| 43 | GPT-5 (high) | OpenAI | 55% | 2025-08-07 |
| 44 | Kimi K2.7 Code | Moonshot | 54% | 2026-06-12 |
| 45 | Grok 4.7 (xhigh) | xAI | 53% | 2026-09-21 |
| 46 | Gemini 3 Flash Preview | Google DeepMind | 51% | 2025-12-17 |
| 47 | GPT-5.4 mini (xhigh) | OpenAI | 51% | 2026-03-17 |
| 48 | GPT-5 mini (high) | OpenAI | 47% | 2025-08-07 |
| 49 | Inkling Small (xhigh) | Thinking Machines | 46% | 2026-07-15 |
| 50 | DeepSeek v4 Pro (max) | DeepSeek | 45% | 2026-04-24 |

Download all data: https://www.superintelligence.com.au/data/benchmarks.csv