microsoft / RetrievalAttentionLinks
Scalable long-context LLM decoding that leverages sparsity—by treating the KV cache as a vector storage system.
☆104Updated 2 months ago
Alternatives and similar repositories for RetrievalAttention
Users that are interested in RetrievalAttention are comparing it to the libraries listed below
Sorting:
- [SIGMOD 2025] PQCache: Product Quantization-based KVCache for Long Context LLM Inference☆78Updated last month
- ☆153Updated 4 months ago
- [OSDI'24] Serving LLM-based Applications Efficiently with Semantic Variable☆200Updated last year
- [NeurIPS 2024] Efficient LLM Scheduling by Learning to Rank☆65Updated last year
- ☆124Updated last year
- ☆79Updated last month
- InfiniGen: Efficient Generative Inference of Large Language Models with Dynamic KV Cache Management (OSDI'24)☆164Updated last year
- Stateful LLM Serving☆89Updated 8 months ago
- [ICLR 2025] TidalDecode: A Fast and Accurate LLM Decoding with Position Persistent Sparse Attention☆49Updated 4 months ago
- ☆82Updated 7 months ago
- ArkVale: Efficient Generative LLM Inference with Recallable Key-Value Eviction (NIPS'24)☆47Updated 11 months ago
- [NeurIPS 2025] ClusterFusion: Expanding Operator Fusion Scope for LLM Inference via Cluster-Level Collective Primitive☆50Updated 2 months ago
- NEO is a LLM inference engine built to save the GPU memory crisis by CPU offloading☆69Updated 5 months ago
- High performance Transformer implementation in C++.☆142Updated 10 months ago
- A framework for generating realistic LLM serving workloads☆87Updated last month
- Efficient Compute-Communication Overlap for Distributed LLM Inference☆63Updated last month
- [DAC'25] Official implement of "HybriMoE: Hybrid CPU-GPU Scheduling and Cache Management for Efficient MoE Inference"☆93Updated 5 months ago
- 16-fold memory access reduction with nearly no loss☆108Updated 8 months ago
- Accelerating Large-Scale Reasoning Model Inference with Sparse Self-Speculative Decoding☆36Updated this week
- Artifact for "Marconi: Prefix Caching for the Era of Hybrid LLMs" [MLSys '25 Outstanding Paper Award, Honorable Mention]☆47Updated 9 months ago
- ☆142Updated last year
- ☆18Updated 5 months ago
- [NeurIPS'25 Spotlight] Adaptive Attention Sparsity with Hierarchical Top-p Pruning☆69Updated last week
- ☆61Updated last month
- PipeRAG: Fast Retrieval-Augmented Generation via Algorithm-System Co-design (KDD 2025)☆28Updated last year
- ☆83Updated last year
- [ICML 2024] Serving LLMs on heterogeneous decentralized clusters.☆32Updated last year
- An experimentation platform for LLM inference optimisation☆35Updated last year
- A tiny yet powerful LLM inference system tailored for researching purpose. vLLM-equivalent performance with only 2k lines of code (2% of …☆294Updated 5 months ago
- Nex Venus Communication Library☆61Updated 3 weeks ago