HELM AFR - MMLU Clinical Knowledge (Igbo): leaderboard
Metric: EM. Source: crfm.stanford.edu. 22 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 2.0 Flash | 70.19 |
| 2 | Gemini 2.0 Flash Lite | 63.4 |
| 3 | GPT-4o (2024-08-06) | 63.02 |
| 4 | DeepSeek V3 | 62.64 |
| 5 | Claude 3.7 Sonnet (20250219) | 61.13 |
| 6 | Llama 3.1 405B Instruct | 56.98 |
| 7 | GPT-4 Turbo | 56.6 |
| 8 | Llama 3.1 70B Instruct | 55.47 |
| 9 | Llama 3.3 70B Instruct | 53.21 |
| 10 | Gemma 2 27B (IT) | 45.66 |
| 11 | Claude 3.5 Haiku (20241022) | 39.62 |
| 12 | Llama 3.1 8B Instruct | 38.49 |
| 13 | Qwen 2.5 72B Instruct | 38.11 |
| 14 | GPT-3.5 Turbo (0125) | 35.09 |
| 15 | Gemma 2 9B (IT) | 31.7 |
Interactive version: theaggregate.ai/benchmark?slug=helm-afr-mmlu-clinical-knowledge-igbo · How It Works · Data refreshed daily, snapshot 2026-09-08.