HELM Capabilities - Omni-MATH — leaderboard

Metric: Acc. Source: crfm.stanford.edu. 51 models tracked.

Top models

#ModelScore
1O4 Mini (2025-04-16)72.03
2O3 (2025-04-16)71.43
3Kimi K265.4
4Grok 460.3
5Qwen3 235B A22B FP8 Throughput54.8
6Claude Sonnet 4 (20250514)51.25
7Claude Opus 4 (20250514)51.12
8GPT-4.1 Mini49.15
9GPT-4.1 (2025-04-14)47.1
10Grok 346.43
11Gemini 2.0 Flash45.9
12DeepSeek R1 052842.4
13Llama 4 Maverick Instruct FP842.23
14Gemini 2.5 Pro (Preview 03-25)41.6
15Palmyra X541.45

Interactive version: theaggregate.ai/benchmark?slug=helm-capabilities-omni-math · How the rankings work · Data refreshed daily, snapshot 2026-07-22.