HELM AFR - Overall (Bambara): leaderboard
Metric: Mean score. Source: crfm.stanford.edu. 22 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 2.0 Flash | 56.99 |
| 2 | Claude 3.7 Sonnet (20250219) | 54.5 |
| 3 | Gemini 2.0 Flash Lite | 50.07 |
| 4 | DeepSeek V3 | 46 |
| 5 | Claude 3.5 Haiku (20241022) | 39.97 |
| 6 | Llama 3.1 70B Instruct | 39.8 |
| 7 | Qwen 2.5 72B Instruct | 39.66 |
| 8 | Llama 3.3 70B Instruct | 38.56 |
| 9 | GPT-4o (2024-08-06) | 37.81 |
| 10 | Gemma 2 27B (IT) | 37.66 |
| 11 | GPT-3.5 Turbo (0125) | 35.53 |
| 12 | Gemma 2 9B (IT) | 33.94 |
| 13 | Llama 3.1 405B Instruct | 33.16 |
| 14 | Llama 3.1 8B Instruct | 33.03 |
| 15 | Mistral 7B Instruct (v0.3) | 32.87 |
Interactive version: theaggregate.ai/benchmark?slug=helm-afr-overall-bambara · How It Works · Data refreshed daily, snapshot 2026-09-08.