Hermes 2 Pro Llama 3 8B
Nous Research · 2024-04-30 · 8B parameters
Hermes 2 Pro Llama 3 8B is Nous Research's instruction-tuned fine-tune of Meta Llama 3 8B, focused on advanced instruction following, function calling, JSON structured outputs, and multi-turn chat using the ChatML prompt format.
Benchmark scores
| Benchmark | Score |
|---|---|
| SnakeBench Average Apples | 0.2 |
| SnakeBench Best Apples | 1.0 |
| SnakeBench Rating | 11.5 |
| SnakeBench Total Apples | 4.0 |
| SnakeBench Win Rate | 36.0 |