FrontierMath - Tier 4
Updated May 2, 2026·1,470 views
- Problem Count
- 300
- Institution
- Epoch AI
- Category
- Math and Reasoning
- Metrics
- Accuracy
- Language
- English
- Difficulty
- Mixed
Overview
FrontierMath - Tier 4 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 - Tier 4 model rankings and full benchmark leaderboard
Browse the latest scores, model modes, release dates, and parameter sizes for FrontierMath - Tier 4.
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 - Tier 4 Rank
| Rank | Model | License | |||
|---|---|---|---|---|---|
 GPT-5.5 Pro Thinking Level · Extra HighTools | 39.60 | 2026-04-23 | Unknown | Closed | |
GPT-5.5 Pro Thinking Level · Extra High | 39.60 | 2026-04-23 | Unknown | Closed | |
GPT-5.5 Pro Thinking Enabled | 39.60 | 2026-04-23 | Unknown | Closed | |
4 | GPT-5.4 Pro Thinking Enabled | 38.00 | 2026-03-05 | Unknown | Closed |
5 | GPT-5.4 Pro Thinking Level · Extra High | 37.50 | 2026-03-05 | Unknown | Closed |
6 | GPT-5.4 Pro Standard ModeToolsInternet | 37.50 | 2026-03-05 | Unknown | Closed |
7 | GPT-5.5 Thinking EnabledTools | 35.40 | 2026-04-23 | Unknown | Closed |
8 | GPT-5.5 Thinking Level · Extra High | 35.40 | 2026-04-23 | Unknown | Closed |
9 | GPT-5.2 Pro Thinking Enabled | 31.30 | 2025-12-11 | Unknown | Closed |
10 | GPT-5.2 Pro Standard ModeToolsInternet | 31.30 | 2025-12-11 | Unknown | Closed |
11 | GPT-5.4 Thinking Level · Extra High | 27.10 | 2026-03-05 | Unknown | Closed |
12 |  Claude Opus 4.6 Thinking Level · High | 22.90 | 2026-02-05 | Unknown | Closed |
13 | Opus 4.7 Thinking Level · Extra High | 22.90 | 2026-04-16 | Unknown | Closed |
14 | Claude Opus 4.6 Thinking Enabled | 20.80 | 2026-02-05 | Unknown | Closed |
15 | Claude Opus 4.6 Thinking Enabled | 20.80 | 2026-02-05 | Unknown | Closed |
16 |  Gemini 3.0 Pro (Preview 11-2025) Thinking Enabled | 18.80 | 2025-11-18 | Unknown | Closed |
17 | Gemini 3.0 Pro (Preview 11-2025) Standard Mode | 18.80 | 2025-11-18 | Unknown | Closed |
18 | GPT-5.2 Thinking Level · Extra High | 18.80 | 2025-12-11 | Unknown | Closed |
19 | GPT-5.2 Thinking Enabled | 18.80 | 2025-12-11 | Unknown | Closed |
20 | GPT-5.2 Thinking Enabled | 16.70 | 2025-12-11 | Unknown | Closed |
21 | Gemini 3.1 Pro Preview Standard Mode | 16.70 | 2026-02-20 | Unknown | Closed |
22 | Gemini 3.1 Pro Preview Thinking Enabled | 16.70 | 2026-02-20 | Unknown | Closed |
23 | GPT-5-Pro Thinking Enabled | 14.60 | 2025-08-07 | Unknown | Closed |
24 | GPT-5-Pro Standard Mode | 14.60 | 2025-08-07 | Unknown | Closed |
25 | GPT-5.2 Thinking Level · Extra HighTools | 14.60 | 2025-12-11 | Unknown | Closed |
26 | Claude Opus 4.6 Thinking Enabled | 14.60 | 2026-02-05 | Unknown | Closed |
27 |  Muse Spark Thinking Enabled | 14.60 | 2026-04-08 | Unknown | Closed |
28 | Muse Spark Standard Mode | 14.60 | 2026-04-08 | Unknown | Closed |
29 | GPT-5 Thinking Enabled | 12.50 | 2025-08-07 | Unknown | Closed |
30 | GPT-5.1 Thinking EnabledTools | 12.50 | 2025-11-12 | Unknown | Closed |
Scroll to load 50 more