Is FrontierMath Tier 4 still meaningful?

Known problems

• Version 1 needed many fixes, so old scores are not comparable. Version 2 fixed 12 Tier 4 problems.^{[2][2] FrontierMath Tiers 1-3 v2 Epoch AI Primary}
• Version 2 re-scored older models when it launched in June 2026.^{[1][1] Benchmark data bundle Epoch AI Primary}

What to look at instead

• FrontierMath Erdős Status:Still useful Yes. It still leaves almost all of its room at the top, though the set is small.

Where it appears in the report

Chapter 1, Benchmarks

• Figure 1. Each New Benchmark Fills Up Faster Than the Last
• Figure 2. Recent benchmarks went from 50% to 90% in under a year
• Table 2. First model to cross each threshold in “Go deeper: The full saturation timeline (14 benchmarks)”
• Table 5. Reasoning and knowledge benchmarks: what they measure, who maintains them, the top score on 5 November 2026 and the known flaw in “Go deeper: Reasoning and knowledge benchmarks (17)”
• Figure 14. OpenAI and Anthropic share only five headline benchmarks