HELM Arabic - Aratrust - Physical Health: leaderboard
Metric: Exact Match (%). Source: crfm.stanford.edu. 41 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude Opus 4.7 | 95.89 |
| 2 | Qwen 3.5 397B A17B | 95.89 |
| 3 | Qwen 2.5 72B Instruct | 94.52 |
| 4 | GPT-4.1 (2025-04-14) | 94.52 |
| 5 | Llama 4 Maverick Instruct FP8 | 94.52 |
| 6 | Mistral Large 3 | 94.52 |
| 7 | Gemini 2.5 Flash (Thinking) | 94.52 |
| 8 | GPT-5.4 (2026-03-05) | 94.52 |
| 9 | Claude Sonnet 4.6 | 93.15 |
| 10 | DeepSeek V3.1 | 93.15 |
| 11 | AceGPT-v2-70B-Chat | 93.15 |
| 12 | Gemma 4 31B (IT) | 91.78 |
| 13 | Qwen 3 Next 80B A3B Instruct | 91.78 |
| 14 | Gemini 2.5 Flash Lite (Thinking) | 91.78 |
| 15 | Qwen 3.5 9B | 90.41 |
Interactive version: theaggregate.ai/benchmark?slug=helm-arabic-aratrust-physical-health · How It Works · Data refreshed daily, snapshot 2026-09-19.