microsoft / vattention
Dynamic Memory Management for Serving LLMs without PagedAttention
☆222Updated this week
Related projects ⓘ
Alternatives and complementary repositories for vattention
- A low-latency & high-throughput serving engine for LLMs☆231Updated last month
- A fast communication-overlapping library for tensor parallelism on GPUs.☆217Updated last week
- [MLSys'24] Atom: Low-bit Quantization for Efficient and Accurate LLM Serving☆277Updated 4 months ago
- A collection of memory efficient attention operators implemented in the Triton language.☆215Updated 5 months ago
- Disaggregated serving system for Large Language Models (LLMs).☆348Updated 2 months ago
- Since the emergence of chatGPT in 2022, the acceleration of Large Language Model has become increasingly important. Here is a list of pap…☆166Updated this week
- Materials for learning SGLang☆75Updated this week
- ☆79Updated 2 months ago
- High performance Transformer implementation in C++.☆78Updated last month
- Zero Bubble Pipeline Parallelism☆279Updated this week
- ☆162Updated 3 months ago
- An efficient GPU support for LLM inference with x-bit quantization (e.g. FP6,FP5).☆196Updated last week
- Efficient and easy multi-instance LLM serving☆192Updated this week
- Flash-LLM: Enabling Cost-Effective and Highly-Efficient Large Generative Model Inference with Unstructured Sparsity☆177Updated last year
- A Easy-to-understand TensorOp Matmul Tutorial☆287Updated last month
- Analyze the inference of Large Language Models (LLMs). Analyze aspects like computation, storage, transmission, and hardware roofline mod…☆310Updated last month
- A large-scale simulation framework for LLM inference☆271Updated 3 weeks ago
- flash attention tutorial written in python, triton, cuda, cutlass☆194Updated 4 months ago
- [ICML 2024] Quest: Query-Aware Sparsity for Efficient Long-Context LLM Inference☆195Updated last week
- MSCCL++: A GPU-driven communication stack for scalable AI applications☆246Updated this week
- QServe: W4A8KV4 Quantization and System Co-design for Efficient LLM Serving☆433Updated 2 months ago
- FlagGems is an operator library for large language models implemented in Triton Language.☆328Updated this week
- [USENIX ATC '24] Accelerating the Training of Large Language Models using Efficient Activation Rematerialization and Optimal Hybrid Paral…☆46Updated 3 months ago
- ☆121Updated this week
- USP: Unified (a.k.a. Hybrid, 2D) Sequence Parallel Attention for Long Context Transformers Model Training and Inference☆352Updated last week
- Applied AI experiments and examples for PyTorch☆159Updated last week
- Ultra-Fast and Cheaper Long-Context LLM Inference☆194Updated this week
- nnScaler: Compiling DNN models for Parallel Training☆62Updated 2 weeks ago
- paper and its code for AI System☆210Updated 2 months ago
- High-speed GEMV kernels, at most 2.7x speedup compared to pytorch baseline.☆87Updated 3 months ago