NVIDIA / kvpress
LLM KV cache compression made easy
☆168Updated this week
Related projects ⓘ
Alternatives and complementary repositories for kvpress
- Fast Matrix Multiplications for Lookup Table-Quantized LLMs☆187Updated this week
- Cataloging released Triton kernels.☆138Updated 2 months ago
- Applied AI experiments and examples for PyTorch☆168Updated 3 weeks ago
- This repository contains the experimental PyTorch native float8 training UX☆212Updated 3 months ago
- ☆96Updated 2 months ago
- Simple and fast low-bit matmul kernels in CUDA / Triton☆147Updated this week
- Fast Inference of MoE Models with CPU-GPU Orchestration☆173Updated this week
- ☆153Updated this week
- Collection of kernels written in Triton language☆68Updated 3 weeks ago
- GEAR: An Efficient KV Cache Compression Recipefor Near-Lossless Generative Inference of LLM☆149Updated 4 months ago
- [ICML 2024] Quest: Query-Aware Sparsity for Efficient Long-Context LLM Inference☆203Updated 3 weeks ago
- Triton-based implementation of Sparse Mixture of Experts.☆185Updated last month
- [ICML 2024] KIVI: A Tuning-Free Asymmetric 2bit Quantization for KV Cache☆243Updated last month
- ☆189Updated 6 months ago
- ring-attention experiments☆97Updated last month
- Breaking Throughput-Latency Trade-off for Long Sequences with Speculative Decoding☆79Updated last week
- [NeurIPS 2024] KVQuant: Towards 10 Million Context Length LLM Inference with KV Cache Quantization☆305Updated 3 months ago
- [MLSys'24] Atom: Low-bit Quantization for Efficient and Accurate LLM Serving☆278Updated 4 months ago
- Code for Palu: Compressing KV-Cache with Low-Rank Projection☆57Updated last week
- 🚀 Efficiently (pre)training foundation models with native PyTorch features, including FSDP for training and SDPA implementation of Flash…☆194Updated this week
- Boosting 4-bit inference kernels with 2:4 Sparsity☆51Updated 2 months ago
- KV cache compression for high-throughput LLM inference☆87Updated last week
- Official repository for LightSeq: Sequence Level Parallelism for Distributed Training of Long Context Transformers☆195Updated 3 months ago
- A safetensors extension to efficiently store sparse quantized tensors on disk☆51Updated this week
- extensible collectives library in triton☆72Updated 2 months ago
- Explorations into some recent techniques surrounding speculative decoding☆212Updated last year
- ShiftAddLLM: Accelerating Pretrained LLMs via Post-Training Multiplication-Less Reparameterization☆87Updated last month
- ☆88Updated 2 months ago
- A subset of PyTorch's neural network modules, written in Python using OpenAI's Triton.☆483Updated 3 weeks ago
- Materials for learning SGLang☆110Updated this week