Llama 3.2 3B Instruct
A highly efficient dense small model by Qwen
Text Models
Developed by
Meta
Optimization
Pruning, distillation, SFT, RLHF
Model family
Llama 3.2
Hardware
H100-80GB (TDP of 700W) type hardware, 8GB+ VRAM recommended.
Use case
API
REST, SDK
Version
3.2
License
https://www.llama.com/llama3_2/license/
Size
3.21B Parameters
Example usage
BUILD SMARTER