HPMLL / BurstGPT
A ChatGPT(GPT-3.5) & GPT-4 Workload Trace to Optimize LLM Serving Systems
☆142Updated 3 months ago
Alternatives and similar repositories for BurstGPT:
Users that are interested in BurstGPT are comparing it to the libraries listed below
- ☆80Updated 2 months ago
- LLM serving cluster simulator☆89Updated 8 months ago
- AlpaServe: Statistical Multiplexing with Model Parallelism for Deep Learning Serving (OSDI 23)☆80Updated last year
- A resilient distributed training framework☆88Updated 9 months ago
- nnScaler: Compiling DNN models for Parallel Training☆87Updated last week
- SpotServe: Serving Generative Large Language Models on Preemptible Instances☆109Updated 10 months ago
- Compiler for Dynamic Neural Networks☆44Updated last year
- ☆61Updated 2 years ago
- ☆15Updated 7 months ago
- High performance Transformer implementation in C++.☆98Updated this week
- Artifact of OSDI '24 paper, ”Llumnix: Dynamic Scheduling for Large Language Model Serving“☆59Updated 7 months ago
- ☆48Updated 7 months ago
- Since the emergence of chatGPT in 2022, the acceleration of Large Language Model has become increasingly important. Here is a list of pap…☆205Updated 3 weeks ago
- This repository is established to store personal notes and annotated papers during daily research.☆106Updated last week
- Bamboo is a system for running large pipeline-parallel DNNs affordably, reliably, and efficiently using spot instances.☆48Updated 2 years ago
- Artifacts for our ASPLOS'23 paper ElasticFlow☆53Updated 8 months ago
- An interference-aware scheduler for fine-grained GPU sharing☆119Updated 8 months ago
- InfiniGen: Efficient Generative Inference of Large Language Models with Dynamic KV Cache Management (OSDI'24)☆100Updated 6 months ago
- Stateful LLM Serving☆44Updated 5 months ago
- Curated collection of papers in MoE model inference☆35Updated this week
- ☆70Updated 3 years ago
- An experimental parallel training platform☆54Updated 9 months ago
- ☆68Updated 2 weeks ago
- ☆13Updated 8 months ago
- Official repository for the paper DynaPipe: Optimizing Multi-task Training through Dynamic Pipelines☆17Updated last year
- [ICML 2024] Serving LLMs on heterogeneous decentralized clusters.☆16Updated 8 months ago
- Here are my personal paper reading notes (including cloud computing, resource management, systems, machine learning, deep learning, and o…☆61Updated last week
- Chimera: Efficiently Training Large-Scale Neural Networks with Bidirectional Pipelines.☆52Updated last year
- ☆16Updated 6 months ago
- Curated collection of papers in machine learning systems☆214Updated 3 weeks ago