ByteDance-Seed / decoupleQLinks
A quantization algorithm for LLM
☆141Updated 11 months ago
Alternatives and similar repositories for decoupleQ
Users that are interested in decoupleQ are comparing it to the libraries listed below
Sorting:
- QQQ is an innovative and hardware-optimized W4A8 quantization solution for LLMs.☆124Updated last month
- Official PyTorch implementation of "FlatQuant: Flatness Matters for LLM Quantization"☆133Updated last week
- An easy-to-use package for implementing SmoothQuant for LLMs☆99Updated last month
- [MLSys'24] Atom: Low-bit Quantization for Efficient and Accurate LLM Serving☆305Updated 11 months ago
- A collection of memory efficient attention operators implemented in the Triton language.☆271Updated 11 months ago
- ☆138Updated 3 months ago
- An efficient GPU support for LLM inference with x-bit quantization (e.g. FP6,FP5).☆251Updated 7 months ago
- [ACL 2024] A novel QAT with Self-Distillation framework to enhance ultra low-bit LLMs.☆114Updated last year
- ☆138Updated last year
- This repository contains integer operators on GPUs for PyTorch.☆205Updated last year
- An algorithm for weight-activation quantization (W4A4, W4A8) of LLMs, supporting both static and dynamic quantization☆136Updated last week
- ☆85Updated 2 months ago
- ☆95Updated 8 months ago
- [ICLR 2025] COAT: Compressing Optimizer States and Activation for Memory-Efficient FP8 Training☆203Updated 2 weeks ago
- Performance of the C++ interface of flash attention and flash attention v2 in large language model (LLM) inference scenarios.☆37Updated 3 months ago
- ☆127Updated 5 months ago
- Code repo for the paper "LLM-QAT Data-Free Quantization Aware Training for Large Language Models"☆287Updated 3 months ago
- Flash-LLM: Enabling Cost-Effective and Highly-Efficient Large Generative Model Inference with Unstructured Sparsity☆211Updated last year
- Code Repository of Evaluating Quantized Large Language Models☆123Updated 8 months ago
- ☆193Updated 3 weeks ago
- ☆146Updated 10 months ago
- 📚FFPA(Split-D): Extend FlashAttention with Split-D for large headdim, O(1) GPU SRAM complexity, 1.8x~3x↑🎉 faster than SDPA EA.☆183Updated 3 weeks ago
- PyTorch bindings for CUTLASS grouped GEMM.☆125Updated 5 months ago
- PyTorch bindings for CUTLASS grouped GEMM.☆93Updated this week
- ☆148Updated 4 months ago
- High-speed GEMV kernels, at most 2.7x speedup compared to pytorch baseline.☆109Updated 10 months ago
- [NeurIPS 2024] KVQuant: Towards 10 Million Context Length LLM Inference with KV Cache Quantization☆355Updated 9 months ago
- Official implementation of the ICLR 2024 paper AffineQuant☆26Updated last year
- Code for Neurips24 paper: QuaRot, an end-to-end 4-bit inference of large language models.☆390Updated 6 months ago
- Code implementation of GPTQv2 (https://arxiv.org/abs/2504.02692)☆45Updated 2 weeks ago