Math benchmark

# FrontierMath Tier 4 leaderboard

> FrontierMath Tier 4 results for 63 AI models, led by GPT-6.1 Sol at 100%. What the benchmark measures, who runs it, and a source for every score.
- Canonical page: https://noometry.com/benchmarks/frontiermath-tier-4
- Last updated: 2026-10-10
- Title: FrontierMath Tier 4 Leaderboard (October 2026): Scores by Model

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 October 10, 2026

## About FrontierMath Tier 4

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

- **Category:** [Math](https://noometry.com/best/math)
- **Introduced:** 2025
- **Format:** Exact answer
- **Unit:** Percent (random guessing ≈ 0%)
- **Official site:** [epoch.ai](https://epoch.ai/frontiermath)

## Top 15 models

Top models on FrontierMath Tier 4

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%
16.  406080100

Sponsored placements are available on pages like this one. [Advertise on Noometry](https://noometry.com/advertise)

## All results

FrontierMath Tier 4 results by model
| # | Model | Provider | Score | Setting | Source | Date |
| --- | --- | --- | --- | --- | --- | --- |
| 1 | [GPT-6.1 Sol](https://noometry.com/models/gpt-6-1-sol) | [OpenAI](https://noometry.com/providers/openai) | 100% | max | [Epoch AI](https://epoch.ai/benchmarks) | 2026-09-29 |
| 2 | [GPT-6 Astra](https://noometry.com/models/gpt-6-astra) | [OpenAI](https://noometry.com/providers/openai) | 97.6% | high | [Epoch AI](https://epoch.ai/benchmarks) | 2026-08-30 |
| 3 | [Claude Opus 5.5](https://noometry.com/models/claude-opus-5-5) | [Anthropic](https://noometry.com/providers/anthropic) | 95% | max | [Epoch AI](https://epoch.ai/benchmarks) | 2026-09-22 |
| 4 | [Claude Fable 5](https://noometry.com/models/claude-fable-5) | [Anthropic](https://noometry.com/providers/anthropic) | 90.2% | max | [Epoch AI](https://epoch.ai/benchmarks) | 2026-06-09 |
| 5 | [GPT-6 Sol](https://noometry.com/models/gpt-6-sol) | [OpenAI](https://noometry.com/providers/openai) | 90% | max | [Epoch AI](https://epoch.ai/benchmarks) | 2026-09-22 |
| 6 | [Claude Fable 5.1](https://noometry.com/models/claude-fable-5-1) | [Anthropic](https://noometry.com/providers/anthropic) | 87.8% | max | [Epoch AI](https://epoch.ai/benchmarks) | 2026-09-01 |
| 7 | [GPT-5.6 Sol](https://noometry.com/models/gpt-5-6-sol) | [OpenAI](https://noometry.com/providers/openai) | 82.9% | max | [Epoch AI](https://epoch.ai/benchmarks) | 2026-07-09 |
| 8 | [Claude Sonnet 5.5](https://noometry.com/models/claude-sonnet-5-5) | [Anthropic](https://noometry.com/providers/anthropic) | 80.5% | max | [Epoch AI](https://epoch.ai/benchmarks) | 2026-09-29 |
| 9 | [GPT-5.5 Pro](https://noometry.com/models/gpt-5-5-pro) | [OpenAI](https://noometry.com/providers/openai) | 78% | xhigh | [Epoch AI](https://epoch.ai/benchmarks) | 2026-06-12 |
| 10 | [AI Co-Mathematician](https://noometry.com/models/ai-co-mathematician) |  [![](/logos/google.svg) Google](https://noometry.com/providers/google) | 75.6% |  | [Epoch AI](https://epoch.ai/benchmarks) | 2026-06-12 |
| 11 | [Claude Opus 5](https://noometry.com/models/claude-opus-5) | [Anthropic](https://noometry.com/providers/anthropic) | 73.2% | max | [Epoch AI](https://epoch.ai/benchmarks) | 2026-07-24 |
| 12 | [GPT-5.5](https://noometry.com/models/gpt-5-5) | [OpenAI](https://noometry.com/providers/openai) | 72.5% | xhigh | [Epoch AI](https://epoch.ai/benchmarks) | 2026-06-11 |
| 13 | [GPT-5.6 Terra](https://noometry.com/models/gpt-5-6-terra) | [OpenAI](https://noometry.com/providers/openai) | 70.7% | max | [Epoch AI](https://epoch.ai/benchmarks) | 2026-07-09 |
| 14 | [GPT-5.6 Luna](https://noometry.com/models/gpt-5-6-luna) | [OpenAI](https://noometry.com/providers/openai) | 61% | max | [Epoch AI](https://epoch.ai/benchmarks) | 2026-07-09 |
| 15 | [GPT-5.4 Pro](https://noometry.com/models/gpt-5-4-pro) | [OpenAI](https://noometry.com/providers/openai) | 58.5% | xhigh | [Epoch AI](https://epoch.ai/benchmarks) | 2026-06-13 |
| 16 | [Claude Opus 4.8](https://noometry.com/models/claude-opus-4-8) | [Anthropic](https://noometry.com/providers/anthropic) | 56.1% | max | [Epoch AI](https://epoch.ai/benchmarks) | 2026-06-10 |
| 17 | [GPT-6 Luna](https://noometry.com/models/gpt-6-luna) | [OpenAI](https://noometry.com/providers/openai) | 56.1% | max | [Epoch AI](https://epoch.ai/benchmarks) | 2026-09-22 |
| 18 | [GPT-5.4](https://noometry.com/models/gpt-5-4) | [OpenAI](https://noometry.com/providers/openai) | 49% | xhigh | [Epoch AI](https://epoch.ai/benchmarks) | 2026-06-11 |
| 19 | [Claude Haiku 5.5](https://noometry.com/models/claude-haiku-5-5) | [Anthropic](https://noometry.com/providers/anthropic) | 46.3% | max | [Epoch AI](https://epoch.ai/benchmarks) | 2026-10-07 |
| 20 | [Muse Spark 1.3](https://noometry.com/models/muse-spark-1-3) |  [![](/logos/meta.svg) Meta](https://noometry.com/providers/meta) | 46.3% | max | [Epoch AI](https://epoch.ai/benchmarks) | 2026-09-18 |
| 21 | [Qwen3.8 Max](https://noometry.com/models/qwen3-8-max) |  [![](/logos/alibaba.svg) Alibaba (Qwen)](https://noometry.com/providers/alibaba) | 46.3% | xhigh | [Epoch AI](https://epoch.ai/benchmarks) | 2026-08-04 |
| 22 | [GPT-5.2 Pro](https://noometry.com/models/gpt-5-2-pro) | [OpenAI](https://noometry.com/providers/openai) | 46% | xhigh | [Epoch AI](https://epoch.ai/benchmarks) | 2026-06-13 |
| 23 | [Kimi K3](https://noometry.com/models/kimi-k3) | [Moonshot AI](https://noometry.com/providers/moonshot) | 39% | max | [Epoch AI](https://epoch.ai/benchmarks) | 2026-07-17 |
| 24 | [Gemini 3.7 Flash](https://noometry.com/models/gemini-3-7-flash) |  [![](/logos/google.svg) Google](https://noometry.com/providers/google) | 36.6% | high | [Epoch AI](https://epoch.ai/benchmarks) | 2026-08-14 |
| 25 | [Qwen3.7 Max](https://noometry.com/models/qwen3-7-max) |  [![](/logos/alibaba.svg) Alibaba (Qwen)](https://noometry.com/providers/alibaba) | 34.1% |  | [Epoch AI](https://epoch.ai/benchmarks) | 2026-06-13 |
| 26 | [Claude Opus 4.7](https://noometry.com/models/claude-opus-4-7) | [Anthropic](https://noometry.com/providers/anthropic) | 31.7% | max | [Epoch AI](https://epoch.ai/benchmarks) | 2026-06-10 |
| 27 | [Grok 4.6](https://noometry.com/models/grok-4-6) | [xAI](https://noometry.com/providers/xai) | 31.7% | xhigh | [Epoch AI](https://epoch.ai/benchmarks) | 2026-08-14 |
| 28 | [GPT-5.2](https://noometry.com/models/gpt-5-2) | [OpenAI](https://noometry.com/providers/openai) | 31.7% | xhigh | [Epoch AI](https://epoch.ai/benchmarks) | 2026-06-11 |
| 29 | [Claude Sonnet 5](https://noometry.com/models/claude-sonnet-5) | [Anthropic](https://noometry.com/providers/anthropic) | 29.3% | max | [Epoch AI](https://epoch.ai/benchmarks) | 2026-06-30 |
| 30 | [GLM-5.2](https://noometry.com/models/glm-5-2) | [Z.ai (Zhipu)](https://noometry.com/providers/zai) | 29.3% | max | [Epoch AI](https://epoch.ai/benchmarks) | 2026-06-19 |
| 31 | [GLM-5.3](https://noometry.com/models/glm-5-3) | [Z.ai (Zhipu)](https://noometry.com/providers/zai) | 29.3% | max | [Epoch AI](https://epoch.ai/benchmarks) | 2026-08-25 |
| 32 | [Claude Opus 4.6](https://noometry.com/models/claude-opus-4-6) | [Anthropic](https://noometry.com/providers/anthropic) | 26.8% | max | [Epoch AI](https://epoch.ai/benchmarks) | 2026-06-11 |
| 33 | [DeepSeek V4.1 Flash](https://noometry.com/models/deepseek-v4-1-flash) |  [![](/logos/deepseek.svg) DeepSeek](https://noometry.com/providers/deepseek) | 26.8% | max | [Epoch AI](https://epoch.ai/benchmarks) | 2026-10-08 |
| 34 | [DeepSeek V4 Pro](https://noometry.com/models/deepseek-v4-pro) |  [![](/logos/deepseek.svg) DeepSeek](https://noometry.com/providers/deepseek) | 26.8% | max | [Epoch AI](https://epoch.ai/benchmarks) | 2026-08-19 |
| 35 | [Gemini 3.1 Pro Preview](https://noometry.com/models/gemini-3-1-pro-preview) |  [![](/logos/google.svg) Google](https://noometry.com/providers/google) | 26.8% |  | [Epoch AI](https://epoch.ai/benchmarks) | 2026-06-11 |
| 36 | [Gemini 3.5 Flash](https://noometry.com/models/gemini-3-5-flash) |  [![](/logos/google.svg) Google](https://noometry.com/providers/google) | 26.8% | high | [Epoch AI](https://epoch.ai/benchmarks) | 2026-06-10 |
| 37 | [Kimi K2.6](https://noometry.com/models/kimi-k2-6) | [Moonshot AI](https://noometry.com/providers/moonshot) | 25.6% |  | [Epoch AI](https://epoch.ai/benchmarks) | 2026-06-10 |
| 38 | [DeepSeek V4 Flash](https://noometry.com/models/deepseek-v4-flash) |  [![](/logos/deepseek.svg) DeepSeek](https://noometry.com/providers/deepseek) | 24.4% | max | [Epoch AI](https://epoch.ai/benchmarks) | 2026-08-02 |
| 39 | [Grok 4.5](https://noometry.com/models/grok-4-5) | [xAI](https://noometry.com/providers/xai) | 24.4% | high | [Epoch AI](https://epoch.ai/benchmarks) | 2026-07-09 |
| 40 | [Gemini 3.6 Flash](https://noometry.com/models/gemini-3-6-flash) |  [![](/logos/google.svg) Google](https://noometry.com/providers/google) | 22% | high | [Epoch AI](https://epoch.ai/benchmarks) | 2026-08-02 |
| 41 | [Gemini 3.8 Flash](https://noometry.com/models/gemini-3-8-flash) |  [![](/logos/google.svg) Google](https://noometry.com/providers/google) | 22% | high | [Epoch AI](https://epoch.ai/benchmarks) | 2026-09-02 |
| 42 | [GPT-5](https://noometry.com/models/gpt-5) | [OpenAI](https://noometry.com/providers/openai) | 22% | high | [Epoch AI](https://epoch.ai/benchmarks) | 2026-06-11 |
| 43 | [GPT-5 Pro](https://noometry.com/models/gpt-5-pro) | [OpenAI](https://noometry.com/providers/openai) | 19.5% | high | [Epoch AI](https://epoch.ai/benchmarks) | 2026-06-12 |
| 44 | [Gemini 3 Flash Preview](https://noometry.com/models/gemini-3-flash-preview) |  [![](/logos/google.svg) Google](https://noometry.com/providers/google) | 17.1% |  | [Epoch AI](https://epoch.ai/benchmarks) | 2026-06-11 |
| 45 | [GLM-5.3-Flash](https://noometry.com/models/glm-5-3-flash) | [Z.ai (Zhipu)](https://noometry.com/providers/zai) | 17.1% | max | [Epoch AI](https://epoch.ai/benchmarks) | 2026-08-27 |
| 46 | [Grok 4.20 (Non-Reasoning)](https://noometry.com/models/grok-4-20) | [xAI](https://noometry.com/providers/xai) | 17.1% |  | [Epoch AI](https://epoch.ai/benchmarks) | 2026-07-13 |
| 47 | [Grok 4.7](https://noometry.com/models/grok-4-7) | [xAI](https://noometry.com/providers/xai) | 17.1% | xhigh | [Epoch AI](https://epoch.ai/benchmarks) | 2026-09-22 |
| 48 | [Inkling-Small](https://noometry.com/models/inkling-small) | [T Thinking Machines Lab](https://noometry.com/providers/thinking-machines) | 17.1% | xhigh | [Epoch AI](https://epoch.ai/benchmarks) | 2026-08-14 |
| 49 | [Grok 4.3](https://noometry.com/models/grok-4-3) | [xAI](https://noometry.com/providers/xai) | 14.6% | high | [Epoch AI](https://epoch.ai/benchmarks) | 2026-06-17 |
| 50 | [GPT-5.4 nano](https://noometry.com/models/gpt-5-4-nano) | [OpenAI](https://noometry.com/providers/openai) | 12.2% | high | [Epoch AI](https://epoch.ai/benchmarks) | 2026-06-12 |
| 51 | [GPT-5 Mini](https://noometry.com/models/gpt-5-mini) | [OpenAI](https://noometry.com/providers/openai) | 12.2% | high | [Epoch AI](https://epoch.ai/benchmarks) | 2026-06-12 |
| 52 | [Kimi K2.7 Code](https://noometry.com/models/kimi-k2-7-code) | [Moonshot AI](https://noometry.com/providers/moonshot) | 12.2% |  | [Epoch AI](https://epoch.ai/benchmarks) | 2026-06-13 |
| 53 | [GPT-5.4 mini](https://noometry.com/models/gpt-5-4-mini) | [OpenAI](https://noometry.com/providers/openai) | 9.8% | xhigh | [Epoch AI](https://epoch.ai/benchmarks) | 2026-06-12 |
| 54 | [Claude Opus 4.5](https://noometry.com/models/claude-opus-4-5) | [Anthropic](https://noometry.com/providers/anthropic) | 4.9% | 32K | [Epoch AI](https://epoch.ai/benchmarks) | 2026-06-11 |
| 55 | [Inkling](https://noometry.com/models/inkling) | [T Thinking Machines Lab](https://noometry.com/providers/thinking-machines) | 4.9% | xhigh | [Epoch AI](https://epoch.ai/benchmarks) | 2026-08-06 |
| 56 | [o4-mini](https://noometry.com/models/o4-mini) | [OpenAI](https://noometry.com/providers/openai) | 4.9% | high | [Epoch AI](https://epoch.ai/benchmarks) | 2026-06-11 |
| 57 | [Claude Opus 4.1](https://noometry.com/models/claude-opus-4-1) | [Anthropic](https://noometry.com/providers/anthropic) | 2.4% | 32K | [Epoch AI](https://epoch.ai/benchmarks) | 2026-06-11 |
| 58 | [Claude Sonnet 4.5](https://noometry.com/models/claude-sonnet-4-5) | [Anthropic](https://noometry.com/providers/anthropic) | 2.4% | 32K | [Epoch AI](https://epoch.ai/benchmarks) | 2026-06-11 |
| 59 | [GPT-5.5 Instant](https://noometry.com/models/gpt-5-5-instant) | [OpenAI](https://noometry.com/providers/openai) | 2.4% |  | [Epoch AI](https://epoch.ai/benchmarks) | 2026-08-02 |
| 60 | [GPT-5 Nano](https://noometry.com/models/gpt-5-nano) | [OpenAI](https://noometry.com/providers/openai) | 2.4% | high | [Epoch AI](https://epoch.ai/benchmarks) | 2026-06-12 |
| 61 | [Gemini 2.5 Pro](https://noometry.com/models/gemini-2-5-pro) |  [![](/logos/google.svg) Google](https://noometry.com/providers/google) | 0% |  | [Epoch AI](https://epoch.ai/benchmarks) | 2026-06-11 |
| 62 | [Gemini 3.5 Flash Lite](https://noometry.com/models/gemini-3-5-flash-lite) |  [![](/logos/google.svg) Google](https://noometry.com/providers/google) | 0% | high | [Epoch AI](https://epoch.ai/benchmarks) | 2026-08-02 |
| 63 | [o3-mini](https://noometry.com/models/o3-mini) | [OpenAI](https://noometry.com/providers/openai) | 0% | high | [Epoch AI](https://epoch.ai/benchmarks) | 2026-06-11 |

## Compare the leaders

-   [GPT-6.1 Sol vs GPT-6 Astra](https://noometry.com/compare/gpt-6-1-sol-vs-gpt-6-astra)
-   [GPT-6.1 Sol vs Claude Opus 5.5](https://noometry.com/compare/claude-opus-5-5-vs-gpt-6-1-sol)
-   [GPT-6.1 Sol vs Claude Fable 5](https://noometry.com/compare/claude-fable-5-vs-gpt-6-1-sol)
-   [GPT-6.1 Sol vs GPT-6 Sol](https://noometry.com/compare/gpt-6-1-sol-vs-gpt-6-sol)
-   [GPT-6 Astra vs Claude Opus 5.5](https://noometry.com/compare/claude-opus-5-5-vs-gpt-6-astra)
-   [GPT-6 Astra vs Claude Fable 5](https://noometry.com/compare/claude-fable-5-vs-gpt-6-astra)

## Other math benchmarks

-   [FrontierMath (Tiers 1-3)](https://noometry.com/benchmarks/frontiermath)
-   [MathArena Final-Answer Competitions](https://noometry.com/benchmarks/matharena)
-   [OTIS Mock AIME 2024-2025](https://noometry.com/benchmarks/otis-mock-aime)
-   [ProofBench](https://noometry.com/benchmarks/proofbench)
-   [Omni-MATH](https://noometry.com/benchmarks/omni-math)
-   [LMArena Math](https://noometry.com/benchmarks/arena-math)
-   [LiveBench Math](https://noometry.com/benchmarks/livebench-math)
-   [MATH Level 5](https://noometry.com/benchmarks/math-level-5)
-   [FrontierMath (Feb 2025 set)](https://noometry.com/benchmarks/frontiermath-2025-02) (reference)
-   [FrontierMath Erdős](https://noometry.com/benchmarks/frontiermath-erdos) (reference)
-   [FrontierMath Tier 4 (v1)](https://noometry.com/benchmarks/frontiermath-tier-4-v1) (reference)
-   [GSM8K](https://noometry.com/benchmarks/gsm8k) (reference)

## 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.

### Cite this page

Noometry. (2026). FrontierMath Tier 4 leaderboard. Retrieved October 10, 2026, from https://noometry.com/benchmarks/frontiermath-tier-4

Quote Noometry with a link back to this page. It is also available in [Markdown](https://noometry.com/md/benchmarks/frontiermath-tier-4.md).