Vahe1994/SpQR

PythonC++
This is stars and forks stats for /Vahe1994/SpQR repository. As of 12 May, 2024 this repository has 451 stars and 33 forks.

SpQR model compression Note: This repository contains quantization algorithm and the model evaluation code for SpQR method for LLM compression; The efficient inference code will be added soon. It accompanies the research paper "SpQR: A Sparse-Quantized Representation for Near-Lossless LLM Weight Compression" . Installation Packages To run SpQR with falcon make sure that you have torch>=2.0.0 with CUDA support. Install packages from requirements.txt: pip install -r requirements.txt Note: the results...
Read on GithubGithub Stats Page
repotechsstarsweeklyforksweekly
TigerResearch/TigerBotPythonOther1.8k01740
deepmind/alphadevPythonC++Starlark6280590
Ma-Lab-Berkeley/CRATEPython5800520
win3zz/CVE-2023-25157Python1300350
DAMO-NLP-SG/Video-LLaMAPython1.7k01390
mikel-brostrom/yolo_trackingPythonDockerfile5.4k01.6k0
hydro-project/hydroflowRustPythonOther3460260
antoinemadec/coc-fzfVim ScriptPythonShell3900270
huchenlei/sd-webui-openpose-editorVueTypeScriptPython3350350
cirosantilli/x86-assembly-cheatAssemblyCMakefile83401200