IST-DASLab / Sparse-MarlinLinks
Boosting 4-bit inference kernels with 2:4 Sparsity
☆82Updated last year
Alternatives and similar repositories for Sparse-Marlin
Users that are interested in Sparse-Marlin are comparing it to the libraries listed below
Sorting:
- [ICLR2025] Breaking Throughput-Latency Trade-off for Long Sequences with Speculative Decoding☆124Updated 9 months ago
- GEAR: An Efficient KV Cache Compression Recipefor Near-Lossless Generative Inference of LLM☆165Updated last year
- ☆141Updated 6 months ago
- Triton-based implementation of Sparse Mixture of Experts.☆238Updated 2 weeks ago
- PyTorch bindings for CUTLASS grouped GEMM.☆113Updated 3 months ago
- [MLSys'24] Atom: Low-bit Quantization for Efficient and Accurate LLM Serving☆319Updated last year
- ☆110Updated last year
- An efficient GPU support for LLM inference with x-bit quantization (e.g. FP6,FP5).☆263Updated last month
- QUICK: Quantization-aware Interleaving and Conflict-free Kernel for efficient LLM inference☆118Updated last year
- 16-fold memory access reduction with nearly no loss☆104Updated 5 months ago
- Cold Compress is a hackable, lightweight, and open-source toolkit for creating and benchmarking cache compression methods built on top of…☆145Updated last year
- Official repository for DistFlashAttn: Distributed Memory-efficient Attention for Long-context LLMs Training☆216Updated last year
- [NeurIPS 2024] KVQuant: Towards 10 Million Context Length LLM Inference with KV Cache Quantization☆377Updated last year
- KV cache compression for high-throughput LLM inference☆136Updated 7 months ago
- [ICLR'25] Fast Inference of MoE Models with CPU-GPU Orchestration☆230Updated 9 months ago
- Repository for Sparse Finetuning of LLMs via modified version of the MosaicML llmfoundry☆42Updated last year
- ☆81Updated 7 months ago
- Ouroboros: Speculative Decoding with Large Model Enhanced Drafting (EMNLP 2024 main)☆109Updated 5 months ago
- ☆88Updated 10 months ago
- Code for data-aware compression of DeepSeek models☆48Updated 3 months ago
- [ICML 2024] KIVI: A Tuning-Free Asymmetric 2bit Quantization for KV Cache☆319Updated 7 months ago
- Fast low-bit matmul kernels in Triton☆357Updated this week
- A safetensors extension to efficiently store sparse quantized tensors on disk☆156Updated last week
- QQQ is an innovative and hardware-optimized W4A8 quantization solution for LLMs.☆139Updated 3 weeks ago
- 🚀 Collection of components for development, training, tuning, and inference of foundation models leveraging PyTorch native components.☆209Updated last week
- extensible collectives library in triton☆88Updated 5 months ago
- Flash-LLM: Enabling Cost-Effective and Highly-Efficient Large Generative Model Inference with Unstructured Sparsity☆219Updated last year
- High-speed GEMV kernels, at most 2.7x speedup compared to pytorch baseline.☆114Updated last year
- ☆124Updated 3 months ago
- ☆74Updated 5 months ago