Epoch AI - Frontiermath Tier 4 V2: leaderboard
Tier 4 of Epoch's FrontierMath v2: the research-level problems a working mathematician would need hours to solve. Held separately from tiers 1–3 because it is the part that was supposed to stay hard, and the top of the board has moved from single digits to the high eighties in under a year.
Metric: Score. Source: epoch.ai. 76 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | gpt-6.1-sol_max | 100 | |
| 2 | gpt-6-astra_high | 97.6 | |
| 3 | gpt-6-astra_xhigh | 97.6 | |
| 4 | gpt-6-astra_max | 97.6 | |
| 5 | gpt-6-astra_medium | 97.56 | |
| 6 | claude-opus-5-5_max | 95 | |
| 7 | claude-fable-5_max | 90.2 | |
| 8 | gpt-6-sol_max | 90 | |
| 9 | claude-fable-5-1_max | 87.8 | |
| 10 | gpt-6-astra_low | 87.8 |
Interactive version: theaggregate.ai/benchmark?slug=epoch-ai-frontiermath-tier-4-v2 · How It Works · Data refreshed daily, snapshot 2026-10-20.