EADST

Quick Review: SpQR: A Sparse-Quantized Representation for Near-Lossless LLM Weight Compression

SpQR: A Sparse-Quantized Representation for Near-Lossless Large Language Model Weight Compression

Core Approach:

  • GPTQ without Outliers: Focuses on eliminating outliers during the GPTQ process, enabling more efficient and accurate weight compression for large language models.
相关标签
About Me
XD
Goals determine what you are going to be.
Category
标签云
Plotly LeetCode PyCharm v0.dev 签证 FP32 Pickle Base64 torchinfo Image2Text LLAMA Quantization Distillation NLP Bin Hungarian Gemma TSV 财报 CTC BeautifulSoup Statistics 证件照 ChatGPT Docker Ptyhon OpenCV Augmentation XGBoost BTC Pandas PIP Knowledge mmap ModelScope transformers Logo 报税 C++ SQLite 强化学习 阿里云 Translation Git VSCode TensorFlow Interview 图形思考法 CUDA Qwen 音频 Vmess ONNX Bert Claude llama.cpp Plate 关于博主 FlashAttention Numpy IndexTTS2 多进程 Michelin Markdown RGB 飞书 logger Dataset Linux WAN 云服务器 CEIR QWEN Rebuttal git-lfs API GIT AI Proxy 域名 Github 搞笑 Nginx Animate git Algorithm YOLO tqdm Random EXCEL CC Crawler ResNet-50 tar scipy NameSilo FP16 Data UNIX FastAPI Tensor 公式 CV 版权 Paper Shortcut TensorRT Clash Qwen2 Datetime Land FP64 Python VPN GoogLeNet v2ray Input UI MD5 Django Agent Disk Food hf Heatmap diffusers 腾讯云 Hilton Permission Search Quantize Streamlit DeepSeek GPTQ Conda SAM Google Excel Video Card Attention Llama CAM Bipartite Baidu PDB Tiktoken Password 净利润 HaggingFace 图标 Paddle 第一性原理 uwsgi RAR 顶会 OCR uWSGI SPIE News Pytorch HuggingFace NLTK 递归学习法 InvalidArgumentError CSV SQL Review Domain FP8 Miniforge TTS OpenAI COCO Sklearn CLAP Firewall PDF VGG-16 LaTeX GPT4 Zip 多线程 Mixtral Bitcoin XML Magnet Qwen2.5 DeepStream Transformers Safetensors 继承 Breakpoint Cloudreve GGML Windows WebCrawler 算法题 Template LoRA Hotel Web JSON Use Website Freesound printf LLM Math BF16 PyTorch Color SVR Jupyter Anaconda Ubuntu icon Jetson Vim Tracking Diagram Pillow
站点统计

本站现有博文323篇,共被浏览801579

本站已经建立2500天!

热门文章
文章归档
回到顶部