All benchmarksexact match (higher is better)
MATH Lvl 5 (4-Shot), exact match
As the model cards publish them, each model’s best result. Facthory has not run them; results are self-reported unless the hub verified them.
| # | Model | Result | Source |
|---|---|---|---|
| 1 | Falcon3-10B-Instruct tiiuae | 25.91 | Open LLM Leaderboard |
| 2 | Falcon3-10B-Base tiiuae | 24.77 | Open LLM Leaderboard |
| 3 | falcon-mamba-7b tiiuae | 3.63 | Open LLM Leaderboard |
