benchmarks.site

Public leaderboard · Math

MathArena

Tests AI models on fresh, hard math problems taken from new competitions and research papers, so the answers cannot have been memorised.

Leaderboard

Top 3 right now

  1. 1stGemini 4 Carbon (high)94.28% ± 3.64%
  2. 2ndGPT-6.1 Sol (max)92.81% ± 3.10%
  3. 3rdGemini 4 Argon (high)84.39% ± 4.65%

What it measures

Models are tested on new math competitions (such as AIME, HMMT and olympiads) shortly after they take place, scored by final-answer accuracy or human grading of proofs.

How to read the numbers

Accuracy on ArXivMath (questions from recent research papers), % with 95% CI — higher is better.

Results

8 rows · higher is better (Accuracy (± 95% CI))CSV
1Gemini 4 Carbon (high)Google94.28% ± 3.64%$5.08
2GPT-6.1 Sol (max)OpenAI92.81% ± 3.10%$0.96
3Gemini 4 Argon (high)Google84.39% ± 4.65%$3.74
4Claude-Opus-5.5 (high)Anthropic82.61% ± 4.72%$1.93
5Kimi K3.1 Pro (max)Moonshot AI58.27% ± 6.84%$2.94
6DeepSeek-V4.1-Flash-1017 (max, DSH)DeepSeek42.48% ± 6.32%$0.33
7Inkling 2 Small (0.99)Thinking Machines Lab31.88% ± 5.14%$0.60
8Muse Spark Open (xhigh)Meta AI27.02% ± 5.61%$1.03

Note: Table is the ArXivMath overall board (most recently updated). MathArena has many other boards (final-answer comps, proof comps, Project Euler, etc.).

Source: MathArena (SRI Lab ETH Zurich / INSAIT) · matharena.ai · results as of 2026-11-13 · retrieved 2026-11-17

Markdown · More Math