Boosting 4-bit inference kernels with 2:4 Sparsity
☆95Sep 4, 2024Updated last year
Alternatives and similar repositories for Sparse-Marlin
Users that are interested in Sparse-Marlin are comparing it to the libraries listed below. We may earn a commission when you buy through links labeled 'Ad' on this page.
Sorting:
- FP16xINT4 LLM inference kernel that can achieve near-ideal ~4x speedups up to medium batchsizes of 16-32 tokens.☆1,061Sep 4, 2024Updated last year
- QQQ is an innovative and hardware-optimized W4A8 quantization solution for LLMs.☆155Aug 21, 2025Updated 8 months ago
- Repository for Sparse Finetuning of LLMs via modified version of the MosaicML llmfoundry☆43Jan 15, 2024Updated 2 years ago
- ☆20Sep 28, 2024Updated last year
- [ICML 2024] KIVI: A Tuning-Free Asymmetric 2bit Quantization for KV Cache☆387Nov 20, 2025Updated 5 months ago
- Deploy to Railway using AI coding agents - Free Credits Offer • AdUse Claude Code, Codex, OpenCode, and more. Autonomous software development now has the infrastructure to match with Railway.
- Fast Matrix Multiplications for Lookup Table-Quantized LLMs☆391Apr 13, 2025Updated last year
- [MLSys'24] Atom: Low-bit Quantization for Efficient and Accurate LLM Serving☆337Jul 2, 2024Updated last year
- ☆105Sep 9, 2024Updated last year
- A high-throughput and memory-efficient inference and serving engine for LLMs☆266Dec 4, 2025Updated 4 months ago
- QUICK: Quantization-aware Interleaving and Conflict-free Kernel for efficient LLM inference☆121Mar 6, 2024Updated 2 years ago
- ☆87Jan 23, 2025Updated last year
- Fast low-bit matmul kernels in Triton☆445Updated this week
- ☆210May 5, 2025Updated 11 months ago
- Work in progress.☆79Nov 25, 2025Updated 5 months ago
- Deploy on Railway without the complexity - Free Credits Offer • AdConnect your repo and Railway handles the rest with instant previews. Quickly provision container image services, databases, and storage volumes.
- [MLSys'25] QServe: W4A8KV4 Quantization and System Co-design for Efficient LLM Serving; [MLSys'25] LServe: Efficient Long-sequence LLM Se…☆834Mar 6, 2025Updated last year
- Code for the paper "QMoE: Practical Sub-1-Bit Compression of Trillion-Parameter Models".☆280Nov 3, 2023Updated 2 years ago
- ☆169Jun 22, 2025Updated 10 months ago
- vLLM performance dashboard☆46Apr 26, 2024Updated 2 years ago
- [NeurIPS 2024] KVQuant: Towards 10 Million Context Length LLM Inference with KV Cache Quantization☆420Aug 13, 2024Updated last year
- Easy and Efficient Quantization for Transformers☆206Mar 25, 2026Updated last month
- Several optimization methods of half-precision general matrix vector multiplication (HGEMV) using CUDA core.☆74Sep 8, 2024Updated last year
- D^2-MoE: Delta Decompression for MoE-based LLMs Compression☆82Mar 25, 2025Updated last year
- A safetensors extension to efficiently store sparse quantized tensors on disk☆274Updated this week
- Deploy to Railway using AI coding agents - Free Credits Offer • AdUse Claude Code, Codex, OpenCode, and more. Autonomous software development now has the infrastructure to match with Railway.
- An easy-to-use package for implementing SmoothQuant for LLMs☆111Apr 7, 2025Updated last year
- ☆22Apr 17, 2025Updated last year
- Fast Hadamard transform in CUDA, with a PyTorch interface☆310Mar 10, 2026Updated last month
- High-speed GEMV kernels, at most 2.7x speedup compared to pytorch baseline.☆128Jul 13, 2024Updated last year
- ☆16Nov 10, 2025Updated 5 months ago
- ☆164Feb 15, 2025Updated last year
- Transformers-compatible library for applying various compression algorithms to LLMs for optimized deployment with vLLM☆3,169Updated this week
- [ICML 2024] SqueezeLLM: Dense-and-Sparse Quantization☆718Aug 13, 2024Updated last year
- HALO: Hadamard-Assisted Low-Precision Optimization and Training method for finetuning LLMs. 🚀 The official implementation of https://arx…☆28Feb 17, 2025Updated last year
- Managed Kubernetes at scale on DigitalOcean • AdDigitalOcean Kubernetes includes the control plane, bandwidth allowance, container registry, automatic updates, and more for free.
- An efficient GPU support for LLM inference with x-bit quantization (e.g. FP6,FP5).☆280Jul 16, 2025Updated 9 months ago
- Code repo for the paper "SpinQuant LLM quantization with learned rotations"☆390Feb 14, 2025Updated last year
- [WIP] Better (FP8) attention for Hopper☆33Feb 24, 2025Updated last year
- SLiM: One-shot Quantized Sparse Plus Low-rank Approximation of LLMs (ICML 2025)☆36Nov 28, 2025Updated 5 months ago
- Code for Neurips24 paper: QuaRot, an end-to-end 4-bit inference of large language models.☆505Nov 26, 2024Updated last year
- This repository contains code for the MicroAdam paper.☆21Dec 14, 2024Updated last year
- A throughput-oriented high-performance serving framework for LLMs☆954Mar 29, 2026Updated last month