Skip to content
Yardstick
  1. Benchmarks
  2. FrontierMath (Tiers 1–3)

FrontierMath (Tiers 1–3)

Original, unpublished mathematics problems from advanced undergraduate to early research level.

ReasoningRun by Epoch AIOfficial leaderboard (opens in a new tab)

Top score
93.7%GPT-6 Astra
Median of field
74.7%14 models scored
Data as of
Sep 29, 2026Epoch AI Benchmarking Hub data export

Standings

Bars share one axis from 0% to 100%.

  1. Rank 1
    GPT-6 Astra93.7%

    OpenAI · $10 / $50New

  2. Rank 1
    GPT-6.1 Sol93.7%

    OpenAI · $2 / $10New

  3. Rank 3
    Claude Opus 5.591.2%

    Anthropic · $4 / $20New

  4. Rank 4
    Claude Fable 5.190.2%

    Anthropic · $10 / $50New

  5. Rank 5
    Claude Sonnet 5.588.8%

    Anthropic · $2 / $10New

  6. Rank 6
    GPT-6 Luna79%

    OpenAI · $0.10 / $0.50New

  7. Rank 7
    Qwen3.8-Max74.7%

    Alibaba · $2 / $6

  8. Rank 8
    Muse Spark 1.374.4%

    Meta · $1.25 / $4.25New

  9. Rank 9
    Kimi K372.2%

    Moonshot AI · $3 / $15Open weights

  10. Rank 10
    GLM-5.368.8%

    Z.ai · $1.40 / $4.40NewOpen weights

  11. Rank 11
    Gemini 3.8 Flash68.4%

    Google · $0.75 / $3.75New

  12. Rank 12
    DeepSeek-V4-Pro64.6%

    DeepSeek · $1.32 / $3.96NewOpen weights

  13. Rank 13
    GLM-5.3-Flash55.8%

    Z.ai · $0.15 / $0.50NewOpen weights

  14. Rank 14
    Grok 4.753%

    SpaceXAI · $2 / $6New

Not reported for FrontierMath (Tiers 1–3) · 15

  • ·
    Claude Haiku 5.5—Not reported

    Anthropic · $0.10 / $0.50New

  • ·
    DeepSeek-V4.1-Flash—Not reported

    DeepSeek · $0.30 / $1.20NewOpen weights

  • ·
    Gemini 3.1 Pro—Not reported

    Google · $2 / $12Preview

  • ·
    Gemini 3.5 Flash-Lite—Not reported

    Google · $0.30 / $2.50

  • ·
    Gemini 4 Argon—Not reported

    Google · $2 / $10NewPreview

  • ·
    Gemma 4 31B—Not reported

    Google · Self-hostedOpen weights

  • ·
    Kimi K2.7 Code—Not reported

    Moonshot AI · $0.95 / $4Open weights

  • ·
    MiMo-V2.6-Pro—Not reported

    Xiaomi · $0.435 / $0.87NewOpen weights

  • ·
    MiniMax M3—Not reported

    MiniMax · $0.30 / $1.20Open weights

  • ·
    Mistral Large 4—Not reported

    Mistral AI · $1.36 / $4.18NewPreview

  • ·
    Mistral Medium 3.5—Not reported

    Mistral AI · $1.50 / $7.50Open weights

  • ·
    Mistral Small 4—Not reported

    Mistral AI · $0.15 / $0.60Open weights

  • ·
    Muse Glimmer 30B—Not reported

    Meta · Self-hostedNewOpen weights

  • ·
    Qwen3.8-27B—Not reported

    Alibaba · $0.50 / $3NewOpen weights

  • ·
    Qwen3.8-Flash-Next—Not reported

    Alibaba · $0.15 / $0.47NewOpen weights

What it measures

Mathematical problem solving on expert-written problems with automatically checkable answers that do not appear online.

How it is scored

Percent answered correctly on Epoch AI's private v2 problem set. Epoch runs every model itself; the model submits code that returns the answer.

Read it with care

Saturating fast: v2 was released in June 2026 and the top scores are already near 94%. Epoch discloses that FrontierMath was funded by OpenAI, which has access to part of the problem set.

Other benchmarks

  • Artificial Analysis Intelligence Index
  • Epoch Capabilities Index
  • LiveBench
  • Terminal-Bench 4.0
  • SWE-Bench Pro (V2 Hard)
  • Arena WebDev
  • GDPval-AA
  • Vending-Bench 2
  • ARC-AGI-2
  • ARC-AGI-3
  • MathArena
  • SimpleBench
  • Humanity's Last Exam (Diamond)
  • GPQA Diamond
  • Arena Vision
  • Arena Text
Yardstick

A plain reading of how today’s SI models compare. Scores are taken from each benchmark’s own leaderboard wherever one exists; anything a lab reported about its own model is marked.

Data as of Oct 8, 2026.

Explore

  • Models
  • Compare
  • Benchmarks
  • Value
  • Labs
  • Methodology & sources

Sources

  • Artificial Analysis (opens in a new tab)
  • Arena (opens in a new tab)
  • Epoch AI Benchmarking Hub (opens in a new tab)
  • ARC Prize (opens in a new tab)
  • Terminal-Bench (opens in a new tab)
  • Scale AI Labs (opens in a new tab)

Yardstick is an independent reference and is not affiliated with, sponsored by or endorsed by any lab listed. Product names and logos are trademarks of their respective owners and are used only to identify them. Benchmarks are imperfect proxies: check a model on your own task before you commit to it.

  • Models
  • Compare
  • Benchmarks
  • Value
  • Labs
  • Models
  • Compare
  • Benchmarks
  • Value
  • Labs
  • Models
  • Compare
  • Benchmarks
  • Value
  • Labs