NEO-MLSys25 / NEOLinks
NEO is a LLM inference engine built to save the GPU memory crisis by CPU offloading
☆39Updated last week
Alternatives and similar repositories for NEO
Users that are interested in NEO are comparing it to the libraries listed below
Sorting:
- Stateful LLM Serving☆73Updated 3 months ago
- ☆62Updated last year
- ☆104Updated 7 months ago
- ☆32Updated last year
- ☆121Updated 5 months ago
- A resilient distributed training framework☆95Updated last year
- A lightweight design for computation-communication overlap.☆143Updated last week
- High performance Transformer implementation in C++.☆125Updated 5 months ago
- Scalable long-context LLM decoding that leverages sparsity—by treating the KV cache as a vector storage system.☆54Updated last week
- Artifact for "Marconi: Prefix Caching for the Era of Hybrid LLMs" [MLSys '25 Outstanding Paper Honorable Mention]☆13Updated 3 months ago
- A ChatGPT(GPT-3.5) & GPT-4 Workload Trace to Optimize LLM Serving Systems☆177Updated 8 months ago
- LLM serving cluster simulator☆106Updated last year
- [SIGMOD 2025] PQCache: Product Quantization-based KVCache for Long Context LLM Inference☆56Updated last week
- [OSDI'24] Serving LLM-based Applications Efficiently with Semantic Variable☆164Updated 9 months ago
- ☆13Updated last year
- ArkVale: Efficient Generative LLM Inference with Recallable Key-Value Eviction (NIPS'24)☆40Updated 6 months ago
- FlexFlow Serve: Low-Latency, High-Performance LLM Serving☆42Updated last month
- Thunder Research Group's Collective Communication Library☆37Updated last year
- Medusa: Accelerating Serverless LLM Inference with Materialization [ASPLOS'25]☆23Updated last month
- InfiniGen: Efficient Generative Inference of Large Language Models with Dynamic KV Cache Management (OSDI'24)☆138Updated 11 months ago
- SpotServe: Serving Generative Large Language Models on Preemptible Instances☆123Updated last year
- Artifact of OSDI '24 paper, ”Llumnix: Dynamic Scheduling for Large Language Model Serving“☆61Updated last year
- DeeperGEMM: crazy optimized version☆69Updated last month
- [DAC'25] Official implement of "HybriMoE: Hybrid CPU-GPU Scheduling and Cache Management for Efficient MoE Inference"☆54Updated 2 weeks ago
- ☆53Updated 4 years ago
- ☆21Updated last year
- Dynamic resources changes for multi-dimensional parallelism training☆25Updated 7 months ago
- ☆72Updated 3 years ago
- [NeurIPS 2024] Efficient LLM Scheduling by Learning to Rank☆48Updated 7 months ago
- Open-source implementation for "Helix: Serving Large Language Models over Heterogeneous GPUs and Network via Max-Flow"☆49Updated 7 months ago