HELM Arabic - Alghafa - Multiple Choice Rating Sentiment No Neutral Task: leaderboard
HELM Arabic (release v2.2.0) AlGhafa Multiple Choice Rating Sentiment No Neutral Task task: the sentiment of rated reviews, positive or negative, from the native AlGhafa Arabic multiple-choice benchmark; 1,000 items, exact match. Percent correct, 0-100, higher is better.
Metric: Exact Match (%). Source: crfm.stanford.edu. Saturation forecast: Around February 2027. 41 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Sonnet 4.6 | 86.3 | #94 |
| 2 | Claude Opus 4.7 | 86 | #51 |
| 3 | GPT-4.1 (2025-04-14) | 84.4 | #257 |
| 4 | Llama 4 Maverick Instruct FP8 | 84.1 | #424 |
| 5 | GPT-5.1 (2025-11-13) | 83.4 | #107 |
| 6 | Llama 4 Scout Instruct | 82.8 | #547 |
| 7 | Llama 3.3 70B Instruct | 82.4 | #541 |
| 8 | jais-adapted-7B-chat | 82 | #1160 |
| 9 | Gemini 2.5 Flash Lite (Non-reasoning) | 81.7 | #432 (Gemini 2.5 Flash Lite) |
| 10 | Claude Haiku 4.5 (20251001) | 81.5 | #264 |
| 11 | Mistral Large 2 (Nov) Instruct (2411) | 81.2 | #480 |
| 12 | ALLaM-7B-Instruct-preview | 81 | #724 |
| 13 | jais-adapted-70B-chat | 80.3 | #639 |
| 14 | Gemma 4 31B (IT) | 79.9 | #178 |
| 15 | Qwen 3.5 397B A17B | 79.5 | #154 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=helm-arabic-alghafa-multiple-choice-rating-sentiment-no-neutral-task · How It Works · Data refreshed daily, snapshot 2026-11-05.