EMRB - L1 Basic Measurement: leaderboard
Metric: Accuracy (%). Source: arxiv.org. 13 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5.5 | 95.8 |
| 2 | Grok 4.5 | 93.4 |
| 3 | GLM-5.2 | 92.9 |
| 4 | Gemini 3.5 Flash | 91.9 |
| 5 | Gemini 3.1 Pro (Preview) | 91.7 |
| 6 | Claude Sonnet 4.6 | 91.5 |
| 7 | DeepSeek V4 Flash | 91.2 |
| 8 | DeepSeek V4 Pro | 90.8 |
| 9 | MiniMax-M3 | 87.7 |
| 10 | Claude Opus 4.7 | 84 |
| 11 | MiMo-V2.5-Pro | 83.7 |
| 12 | Kimi K2.6 | 66.4 |
| 13 | Llama 3.3 70B | 52.5 |
Interactive version: theaggregate.ai/benchmark?slug=emrb-l1-basic-measurement · How It Works · Data refreshed daily, snapshot 2026-09-19.