ai.paski.devLlama 3.2 3B Instruct
llmA lightweight 3B-parameter instruction-tuned model created via pruning and distillation from Llama 3.1 8B, designed for on-device and edge deployment.
Metadata
| org | Meta |
| released | 2024-10-24 |
| params | 3B |
| license | llama-3.2 |
| kind | llm |
| hf id | meta-llama/Llama-3.2-3B-Instruct |
Benchmarks
| MMLU | 63.4 | source ↗ |
Lineage
↑ distillLlama 3.1 8B