HELM Arabic - Alghafa - Multiple Choice Rating Sentiment Task: leaderboard
Metric: Exact Match (%). Source: crfm.stanford.edu. 41 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude Sonnet 4.6 | 63.1 |
| 2 | GPT-4.1 (2025-04-14) | 63 |
| 3 | Gemma 4 31B (IT) | 60.2 |
| 4 | Claude Opus 4.7 | 60.2 |
| 5 | Llama 4 Maverick Instruct FP8 | 59.6 |
| 6 | Llama 3.3 70B Instruct | 57.9 |
| 7 | jais-adapted-7B-chat | 56.9 |
| 8 | Llama 4 Scout Instruct | 56.4 |
| 9 | AceGPT-v2-8B-Chat | 56.4 |
| 10 | jais-adapted-70B-chat | 56.3 |
| 11 | Qwen 3.5 397B A17B | 56.1 |
| 12 | Gemini 2.5 Flash (Thinking) | 56 |
| 13 | AceGPT-v2-70B-Chat | 56 |
| 14 | Claude Haiku 4.5 (20251001) | 55.8 |
| 15 | AceGPT-v2-32B-Chat | 55.7 |
Interactive version: theaggregate.ai/benchmark?slug=helm-arabic-alghafa-multiple-choice-rating-sentiment-task · How It Works · Data refreshed daily, snapshot 2026-09-19.