NEO-MLSys25 / NEO
NEO is a LLM inference engine built to save the GPU memory crisis by CPU offloading
☆19Updated last month
Alternatives and similar repositories for NEO:
Users that are interested in NEO are comparing it to the libraries listed below
- Open-source implementation for "Helix: Serving Large Language Models over Heterogeneous GPUs and Network via Max-Flow"☆29Updated 4 months ago
- Stateful LLM Serving☆50Updated 3 weeks ago
- High performance Transformer implementation in C++.☆113Updated 2 months ago
- ☆19Updated 10 months ago
- ☆55Updated 9 months ago
- ☆91Updated 3 months ago
- Medusa: Accelerating Serverless LLM Inference with Materialization [ASPLOS'25]☆19Updated 3 months ago
- ☆70Updated 3 years ago
- Supplemental materials for The ASPLOS 2025 / EuroSys 2025 Contest on Intra-Operator Parallelism for Distributed Deep Learning☆23Updated 3 months ago
- Ultra | Ultimate | Unified CCL☆57Updated last month
- Efficient Interactive LLM Serving with Proxy Model-based Sequence Length Prediction | A tiny BERT model can tell you the verbosity of an …☆33Updated 10 months ago
- ☆16Updated 11 months ago
- ☆36Updated 5 months ago
- ☆53Updated 4 years ago
- Artifacts for our ASPLOS'23 paper ElasticFlow☆52Updated 10 months ago
- FlexFlow Serve: Low-Latency, High-Performance LLM Serving☆34Updated this week
- ☆91Updated 4 months ago
- LLM serving cluster simulator☆95Updated 11 months ago
- Official repository for the paper DynaPipe: Optimizing Multi-task Training through Dynamic Pipelines☆17Updated last year
- ☆14Updated 2 years ago
- Artifact of OSDI '24 paper, ”Llumnix: Dynamic Scheduling for Large Language Model Serving“☆60Updated 10 months ago
- ☆10Updated 5 months ago
- SOTA Learning-augmented Systems☆35Updated 2 years ago
- ☆23Updated 8 months ago
- InfiniGen: Efficient Generative Inference of Large Language Models with Dynamic KV Cache Management (OSDI'24)☆119Updated 8 months ago
- A ChatGPT(GPT-3.5) & GPT-4 Workload Trace to Optimize LLM Serving Systems☆157Updated 5 months ago
- Artifact for "Apparate: Rethinking Early Exits to Tame Latency-Throughput Tensions in ML Serving" [SOSP '24]☆23Updated 4 months ago
- SHADE: Enable Fundamental Cacheability for Distributed Deep Learning Training☆31Updated 2 years ago
- A GPU-accelerated DNN inference serving system that supports instant kernel preemption and biased concurrent execution in GPU scheduling.☆42Updated 2 years ago
- ☆12Updated 9 months ago