FrontierMath
Updated May 2, 2026·1,776 views
- Problem Count
- 300
- Institution
- Epoch AI
- Category
- Math and Reasoning
- Metrics
- Accuracy
- Language
- English
- Difficulty
- Mixed
Overview
FrontierMath is an AI benchmark used to evaluate model capabilities. Review its overview, metrics, official resources, and model leaderboard results on DataLearnerAI.
Related resources
Latest FrontierMath model rankings and full benchmark leaderboard
Browse the latest scores, model modes, release dates, and parameter sizes for FrontierMath.
Source: DataLearnerAI
Data sourced primarily from official releases (GitHub, Hugging Face, papers), then benchmark leaderboards, then third-party evaluators. Learn about our data methodology
Model Mode Legend
License:
Origin:
Model release cutoff:
FrontierMath Rank
| Rank | Model | License | |||
|---|---|---|---|---|---|
 GPT-5.5 Pro Thinking Level · Extra HighTools | 52.40 | 2026-04-23 | Unknown | Closed | |
GPT-5.5 Thinking EnabledTools | 51.70 | 2026-04-23 | Unknown | Closed | |
GPT-5.4 Pro Thinking Enabled | 50.00 | 2026-03-05 | Unknown | Closed | |
4 | GPT-5.4 Pro Thinking Level · Extra High | 50.00 | 2026-03-05 | Unknown | Closed |
5 | GPT-5.4 Thinking Level · Extra High | 47.60 | 2026-03-05 | Unknown | Closed |
6 |  Opus 4.7 Thinking Level · Extra High | 43.80 | 2026-04-16 | Unknown | Closed |
7 | Claude Opus 4.6 Thinking Level · High | 40.70 | 2026-02-05 | Unknown | Closed |
8 | GPT-5.2 Thinking Level · Extra HighTools | 40.30 | 2025-12-11 | Unknown | Closed |
9 |  Muse Spark Thinking Enabled | 39.00 | 2026-04-08 | Unknown | Closed |
10 |  Gemini 3.0 Pro (Preview 11-2025) Thinking Enabled | 38.00 | 2025-11-18 | Unknown | Closed |
11 | Gemini 3.1 Pro Preview Thinking Enabled | 36.90 | 2026-02-20 | Unknown | Closed |
12 | Gemini 2.5 Deep Think Deep Thinking Mode | 29.00 | 2025-08-01 | Unknown | Closed |
13 | GPT-5.1 Thinking EnabledTools | 26.70 | 2025-11-12 | Unknown | Closed |
14 | GPT-5 Thinking EnabledTools | 26.30 | 2025-08-07 | Unknown | Closed |
15 | GPT-5 Thinking Level · Medium | 24.80 | 2025-08-07 | Unknown | Closed |
16 | GPT-5 Thinking Level · High | 24.80 | 2025-08-07 | Unknown | Closed |
17 | Opus 4.5 Extended Thinking | 20.70 | 2025-11-25 | Unknown | Closed |
18 | OpenAI o4 - mini Thinking Level · Medium | 19.30 | 2025-04-16 | Unknown | Closed |
19 | GPT-5-mini Thinking Level · Medium | 19.30 | 2025-08-07 | Unknown | Closed |
20 | GPT-5-mini Thinking Level · High | 19.00 | 2025-08-07 | Unknown | Closed |
21 | OpenAI o4 - mini Thinking Level · High | 17.20 | 2025-04-16 | Unknown | Closed |
22 |  Grok 4 Standard Mode | 12.10 | 2025-07-10 | Unknown | Closed |
23 | OpenAI o3-mini (high) Thinking Level · High | 11.00 | 2025-01-31 | Unknown | Closed |
24 | Gemini 2.5-Pro Standard Mode | 11.00 | 2025-06-05 | Unknown | Closed |
25 | OpenAI o3 Thinking Level · Low | 10.30 | 2025-04-16 | Unknown | Closed |
26 | OpenAI o3 Thinking Level · High | 10.30 | 2025-04-16 | Unknown | Closed |
27 | Gemini-2.5-Pro-Preview-05-06 Standard Mode | 10.30 | 2025-05-06 | Unknown | Closed |
28 | OpenAI o3 Thinking Level · Medium | 10.00 | 2025-04-16 | Unknown | Closed |
29 | OpenAI o4 - mini Thinking Level · Low | 9.70 | 2025-04-16 | Unknown | Closed |
30 | OpenAI o1 Thinking Level · High | 9.30 | 2024-12-05 | Unknown | Closed |
Scroll to load 30 more