HELM AFR - MMLU Clinical Knowledge (Bambara): leaderboard
Metric: EM. Source: crfm.stanford.edu. 22 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 2.0 Flash | 63.4 |
| 2 | Gemini 2.0 Flash Lite | 60 |
| 3 | Claude 3.7 Sonnet (20250219) | 56.98 |
| 4 | DeepSeek V3 | 47.17 |
| 5 | GPT-4o (2024-08-06) | 43.02 |
| 6 | Claude 3.5 Haiku (20241022) | 42.64 |
| 7 | Llama 3.1 70B Instruct | 40.75 |
| 8 | Qwen 2.5 72B Instruct | 40.75 |
| 9 | Llama 3.3 70B Instruct | 39.25 |
| 10 | GPT-3.5 Turbo (0125) | 35.85 |
| 11 | Llama 3.1 8B Instruct | 33.96 |
| 12 | GPT-4 Turbo | 32.83 |
| 13 | Gemma 2 27B (IT) | 30.94 |
| 14 | Mistral 7B Instruct (v0.3) | 27.17 |
| 15 | GPT-2 | 26.79 |
Interactive version: theaggregate.ai/benchmark?slug=helm-afr-mmlu-clinical-knowledge-bambara · How It Works · Data refreshed daily, snapshot 2026-09-08.