microsoft / RetrievalAttentionLinks
Scalable long-context LLM decoding that leverages sparsity—by treating the KV cache as a vector storage system.
☆65Updated last week
Alternatives and similar repositories for RetrievalAttention
Users that are interested in RetrievalAttention are comparing it to the libraries listed below
Sorting:
- ☆106Updated 8 months ago
- [SIGMOD 2025] PQCache: Product Quantization-based KVCache for Long Context LLM Inference☆61Updated 3 weeks ago
- ☆120Updated 6 months ago
- Stateful LLM Serving☆76Updated 4 months ago
- [OSDI'24] Serving LLM-based Applications Efficiently with Semantic Variable☆168Updated 9 months ago
- ☆64Updated last year
- 16-fold memory access reduction with nearly no loss☆100Updated 3 months ago
- ☆110Updated 9 months ago
- [NeurIPS 2024] Efficient LLM Scheduling by Learning to Rank☆49Updated 8 months ago
- ☆54Updated last year
- NEO is a LLM inference engine built to save the GPU memory crisis by CPU offloading☆44Updated 3 weeks ago
- High performance Transformer implementation in C++.☆125Updated 5 months ago
- ☆77Updated 2 months ago
- A lightweight design for computation-communication overlap.☆146Updated 3 weeks ago
- InfiniGen: Efficient Generative Inference of Large Language Models with Dynamic KV Cache Management (OSDI'24)☆142Updated last year
- ArkVale: Efficient Generative LLM Inference with Recallable Key-Value Eviction (NIPS'24)☆41Updated 6 months ago
- Flash-LLM: Enabling Cost-Effective and Highly-Efficient Large Generative Model Inference with Unstructured Sparsity☆216Updated last year
- nnScaler: Compiling DNN models for Parallel Training☆113Updated last week
- PyTorch library for cost-effective, fast and easy serving of MoE models.☆207Updated last week
- ☆32Updated last year
- Modular and structured prompt caching for low-latency LLM inference☆97Updated 8 months ago
- [ICLR 2025] TidalDecode: A Fast and Accurate LLM Decoding with Position Persistent Sparse Attention☆40Updated 2 months ago
- PipeRAG: Fast Retrieval-Augmented Generation via Algorithm-System Co-design (KDD 2025)☆22Updated last year
- ☆60Updated 2 months ago
- PyTorch implementation of paper "Response Length Perception and Sequence Scheduling: An LLM-Empowered LLM Inference Pipeline".☆89Updated 2 years ago
- [ICLR2025] Breaking Throughput-Latency Trade-off for Long Sequences with Speculative Decoding☆121Updated 7 months ago
- SpotServe: Serving Generative Large Language Models on Preemptible Instances☆123Updated last year
- [ICML 2024] Quest: Query-Aware Sparsity for Efficient Long-Context LLM Inference☆303Updated this week
- [ICML 2024] Serving LLMs on heterogeneous decentralized clusters.☆26Updated last year
- DeeperGEMM: crazy optimized version☆69Updated 2 months ago