HELM AFR - Winogrande (Southern Sotho): leaderboard
The "Winogrande (Southern Sotho)" subtask of HELM AFR. Stanford HELM's African-languages project (mmlu-winogrande-afr): MMLU subject exams and Winogrande coreference translated into sixteen African languages, scored against the same items in English so a model's loss from the language switch...
Metric: EM. Source: crfm.stanford.edu. Saturation forecast: Estimated already saturated. 22 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-4o (2024-08-06) | 70.12 | #335 |
| 2 | Claude 3.7 Sonnet (20250219) | 69.38 | #203 |
| 3 | Gemini 2.0 Flash | 68.65 | #344 |
| 4 | GPT-4 Turbo | 67.29 | #406 |
| 5 | DeepSeek V3 | 60.78 | #323 |
| 6 | Gemini 2.0 Flash Lite | 59.76 | #453 |
| 7 | Llama 3.1 405B Instruct | 57.16 | #461 |
| 8 | Llama 3.3 70B Instruct | 54.39 | #536 |
| 9 | Claude 3.5 Haiku (20241022) | 53.76 | #591 |
| 10 | Gemma 2 27B (IT) | 53.31 | #649 |
| 11 | Llama 3.1 70B Instruct | 53.14 | #566 |
| 12 | Mixtral 8x7B Instruct | 52.29 | #1172 |
| 13 | Qwen 2.5 72B Instruct | 51.84 | #458 |
| 14 | Mistral 7B Instruct (v0.3) | 51.05 | #1276 |
| 15 | GPT-2 XL | 49.8 | #1617 |
No result here: #4 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=helm-afr-winogrande-southern-sotho · How It Works · Data refreshed daily, snapshot 2026-10-23.