Model results
FrontierMath
Choose a shared evaluation protocol to compare model scores. Versions, sources and configurations stay separate.
Atlas snapshot captured 10/19/2026, 10:20:28 AM UTC. This is an import timestamp, not the evaluation date. Coverage may be incomplete.
Reviewed recommendation and limits
Use the relevant tier and private/public split for difficult mathematical reasoning.
Each tier, release and public/private split separately. Imported groups below may use different protocols; this review does not certify each run. Difficulty is specialized; tier, dataset release, tools and grader must match before comparing scores.
FrontierMath 2025-02-28 · private
- Source
- Epoch AI Benchmarking Hub
- Version
- 2025-02-28-private
- Measurement date
- Not reported
- Metric
- % · Higher is better
- Evaluation scope
- Epoch AI run · task v1 · effort high · superseded by FrontierMath Tiers 1–3 v2 · private · result date not supplied
Exact score ties share a position. Positions do not establish statistical significance. Missing metadata is shown as unreported; matching imported metadata is not independent verification of the protocol.
#1GPT-5.2
openai/gpt-5.2-20251211
40.3%
View modelEvidence independent#2Gemini 3.5 Flash
google/gemini-3.5-flash-20260519
39%
View modelEvidence independent#3GPT-5
openai/gpt-5-2025-08-07
32.4%
View modelEvidence independent#4GPT-5.1
openai/gpt-5.1-20251113
31%
View modelEvidence independent#5GPT-5.4 Mini
openai/gpt-5.4-mini-20260317
28.3%
View modelEvidence independent#6GPT-5 Mini
openai/gpt-5-mini-2025-08-07
27.2%
View modelEvidence independent#7GPT-5.4 Nano
openai/gpt-5.4-nano-20260317
25.9%
View modelEvidence independent#8o4 Mini
openai/o4-mini-2025-04-16
24.8%
View modelEvidence independent#9o3
openai/o3-2025-04-16
18.7%
View modelEvidence independent#10o3 Mini
openai/o3-mini-2025-01-31
12.4%
View modelEvidence independent#11o1
openai/o1-2024-12-17
9.3%
View modelEvidence independent#12GPT-5 Nano
openai/gpt-5-nano-2025-08-07
8.3%
View modelEvidence independent