HELM VHELM - Mmmu - Geography: leaderboard
Metric: Prefix Quasi-Exact Match (%). Source: crfm.stanford.edu. 48 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | O3 (2025-04-16) | 81.48 |
| 2 | Gemini 2.0 Flash Lite | 74.07 |
| 3 | Claude 3.5 Sonnet (20241022) | 70.37 |
| 4 | GPT-4o (2024-08-06) | 70.37 |
| 5 | O1 (2024-12-17) | 70.37 |
| 6 | GPT-4.5 | 70.37 |
| 7 | Claude 3.5 Sonnet (20240620) | 66.67 |
| 8 | Gemini 2.0 Flash | 66.67 |
| 9 | Claude 3 Opus (20240229) | 66.67 |
| 10 | Gemini 2.0 Pro (Preview 02-05) | 66.67 |
| 11 | GPT-4.1 (2025-04-14) | 62.96 |
| 12 | Gemini 2.5 Pro (Preview 03-25) | 62.96 |
| 13 | Claude 3 Haiku (20240307) | 62.96 |
| 14 | Gemini 1.5 Pro (002) | 59.26 |
| 15 | GPT-4o Mini (2024-07-18) | 59.26 |
Interactive version: theaggregate.ai/benchmark?slug=helm-vhelm-mmmu-geography · How It Works · Data refreshed daily, snapshot 2026-09-19.