HELM Capabilities - Omni-MATH — leaderboard
Metric: Acc. Source: crfm.stanford.edu. 51 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | O4 Mini (2025-04-16) | 72.03 |
| 2 | O3 (2025-04-16) | 71.43 |
| 3 | Kimi K2 | 65.4 |
| 4 | Grok 4 | 60.3 |
| 5 | Qwen3 235B A22B FP8 Throughput | 54.8 |
| 6 | Claude Sonnet 4 (20250514) | 51.25 |
| 7 | Claude Opus 4 (20250514) | 51.12 |
| 8 | GPT-4.1 Mini | 49.15 |
| 9 | GPT-4.1 (2025-04-14) | 47.1 |
| 10 | Grok 3 | 46.43 |
| 11 | Gemini 2.0 Flash | 45.9 |
| 12 | DeepSeek R1 0528 | 42.4 |
| 13 | Llama 4 Maverick Instruct FP8 | 42.23 |
| 14 | Gemini 2.5 Pro (Preview 03-25) | 41.6 |
| 15 | Palmyra X5 | 41.45 |
Interactive version: theaggregate.ai/benchmark?slug=helm-capabilities-omni-math · How the rankings work · Data refreshed daily, snapshot 2026-07-22.