NEO-MLSys25 / NEOLinks
NEO is a LLM inference engine built to save the GPU memory crisis by CPU offloading
☆71Updated 6 months ago
Alternatives and similar repositories for NEO
Users that are interested in NEO are comparing it to the libraries listed below
Sorting:
- ☆65Updated last month
- A framework for generating realistic LLM serving workloads☆93Updated 2 months ago
- Stateful LLM Serving☆90Updated 9 months ago
- Open-source implementation for "Helix: Serving Large Language Models over Heterogeneous GPUs and Network via Max-Flow"☆74Updated 2 months ago
- InfiniGen: Efficient Generative Inference of Large Language Models with Dynamic KV Cache Management (OSDI'24)☆167Updated last year
- ☆125Updated last year
- Medusa: Accelerating Serverless LLM Inference with Materialization [ASPLOS'25]☆40Updated 7 months ago
- ☆156Updated 5 months ago
- ☆79Updated 2 months ago
- A ChatGPT(GPT-3.5) & GPT-4 Workload Trace to Optimize LLM Serving Systems☆224Updated 4 months ago
- High performance Transformer implementation in C++.☆146Updated 11 months ago
- SpotServe: Serving Generative Large Language Models on Preemptible Instances☆133Updated last year
- ☆79Updated 3 years ago
- Here are my personal paper reading notes (including cloud computing, resource management, systems, machine learning, deep learning, and o…☆138Updated last month
- Scalable long-context LLM decoding that leverages sparsity—by treating the KV cache as a vector storage system.☆106Updated 3 months ago
- Artifact of OSDI '24 paper, ”Llumnix: Dynamic Scheduling for Large Language Model Serving“☆64Updated last year
- [SIGMOD 2025] PQCache: Product Quantization-based KVCache for Long Context LLM Inference☆80Updated last week
- LLM serving cluster simulator☆126Updated last year
- A lightweight design for computation-communication overlap.☆196Updated 2 months ago
- ☆15Updated 7 months ago
- NVSHMEM‑Tutorial: Build a DeepEP‑like GPU Buffer☆146Updated 3 months ago
- [NeurIPS 2025] ClusterFusion: Expanding Operator Fusion Scope for LLM Inference via Cluster-Level Collective Primitive☆50Updated last week
- Pie: Programmable LLM Serving☆79Updated this week
- ☆23Updated last year
- Compiler for Dynamic Neural Networks☆46Updated 2 years ago
- ☆41Updated last year
- gLLM: Global Balanced Pipeline Parallelism System for Distributed LLM Serving with Token Throttling☆51Updated this week
- [OSDI'24] Serving LLM-based Applications Efficiently with Semantic Variable☆203Updated last year
- An interference-aware scheduler for fine-grained GPU sharing☆154Updated 3 weeks ago
- Research prototype of PRISM — a cost-efficient multi-LLM serving system with flexible time- and space-based GPU sharing.☆49Updated 4 months ago