Quantizing Llama-3.2-1B Eric Hartford
I am creating several quants of Llama-3.1-1B for the purposes of testing vLLM Marlin.
- https://huggingface.co/QuixiAI/Llama-3.2-1B
- https://huggingface.co/QuixiAI/Llama-3.2-1B-FP8-Dynamic
- https://huggingface.co/QuixiAI/Llama-3.2-1B-MXFP4
- https://huggingface.co/QuixiAI/Llama-3.2-1B-NVFP4A16
- https://huggingface.co/QuixiAI/Llama-3.2-1B-W4A16-AWQ
- https://huggingface.co/QuixiAI/Llama-3.2-1B-W4A16-GPTQ
- https://huggingface.co/QuixiAI/Llama-3.2-1B-W8A16-GPTQ
The script I used to quant this: quant.py
- Downloads last month
- 2
Model tree for QuixiAI/Llama-3.2-1B-W8A16-GPTQ
Base model
meta-llama/Llama-3.2-1B-Instruct