HELM Arabic - Alghafa - Multiple Choice Grounded Statement Xglue Mlqa Task: leaderboard
HELM Arabic (release v2.2.0) AlGhafa Multiple Choice Grounded Statement Xglue Mlqa Task task: choosing the statement supported by an XGLUE MLQA reading passage, from the native AlGhafa Arabic multiple-choice benchmark; 150 items, exact match. Percent correct, 0-100, higher is better.
Metric: Exact Match (%). Source: crfm.stanford.edu. Saturation forecast: Estimated already saturated. 41 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Qwen 2.5 72B Instruct | 96.67 | #469 |
| 2 | Claude Opus 4.7 | 96 | #51 |
| 3 | Llama 4 Scout Instruct | 95.33 | #547 |
| 4 | Claude Sonnet 4.6 | 94.67 | #94 |
| 5 | Mistral Large 3 | 94.67 | #410 |
| 6 | GPT-4.1 (2025-04-14) | 94 | #257 |
| 7 | Qwen 3 Next 80B A3B Instruct | 94 | #381 |
| 8 | Gemini 2.5 Flash Lite (Non-reasoning) | 94 | #432 (Gemini 2.5 Flash Lite) |
| 9 | GPT-5.4 (2026-03-05) | 94 | #63 |
| 10 | Gemma 4 31B (IT) | 93.33 | #178 |
| 11 | GPT-5.1 (2025-11-13) | 93.33 | #107 |
| 12 | Qwen 3.5 9B | 92.67 | #391 |
| 13 | Gemini 2.5 Flash (Non-reasoning) | 92.67 | #255 (Gemini 2.5 Flash) |
| 14 | AceGPT-v2-70B-Chat | 92.67 | #421 |
| 15 | GPT-5.4 Mini (2026-03-17) | 92.67 | #180 |
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-grounded-statement-xglue-mlqa-task · How It Works · Data refreshed daily, snapshot 2026-11-05.