BitStack
/

BitStack-Llama-3-8B

Text Generation

text-generation-inference

8-bit precision

Model card Files Files and versions

This repository contains the BitStack-Llama-3-8B model as presented in BitStack: Fine-Grained Size Control for Compressed Large Language Models in Variable Memory Environments.

Downloads last month: 4

Safetensors

Model size

15.7B params

Tensor type

F16

·

U8

·