timlee0212 / SiDA-MoELinks
Code for MLSys 2024 Paper "SiDA-MoE: Sparsity-Inspired Data-Aware Serving for Efficient and Scalable Large Mixture-of-Experts Models"
☆21Updated last year
Alternatives and similar repositories for SiDA-MoE
Users that are interested in SiDA-MoE are comparing it to the libraries listed below
Sorting:
- ☆69Updated last year
- [OSDI'24] Serving LLM-based Applications Efficiently with Semantic Variable☆181Updated 11 months ago
- Stateful LLM Serving☆81Updated 5 months ago
- [ICML 2024] Serving LLMs on heterogeneous decentralized clusters.☆29Updated last year
- ☆116Updated 9 months ago
- A resilient distributed training framework☆94Updated last year
- ☆131Updated last month
- [NeurIPS 2024] Efficient LLM Scheduling by Learning to Rank☆58Updated 9 months ago
- PyTorch implementation of paper "Response Length Perception and Sequence Scheduling: An LLM-Empowered LLM Inference Pipeline".☆90Updated 2 years ago
- ☆33Updated last year
- ☆123Updated 10 months ago
- kvcached: Elastic KV cache for dynamic GPU sharing and efficient multi-LLM inference.☆66Updated this week
- PyTorch library for cost-effective, fast and easy serving of MoE models.☆224Updated last month
- FlexFlow Serve: Low-Latency, High-Performance LLM Serving☆56Updated this week
- AlpaServe: Statistical Multiplexing with Model Parallelism for Deep Learning Serving (OSDI 23)☆84Updated 2 years ago
- InfiniGen: Efficient Generative Inference of Large Language Models with Dynamic KV Cache Management (OSDI'24)☆152Updated last year
- SpotServe: Serving Generative Large Language Models on Preemptible Instances☆128Updated last year
- A lightweight design for computation-communication overlap.☆160Updated last week
- Explore Inter-layer Expert Affinity in MoE Model Inference☆13Updated last year
- LLM checkpointing for DeepSpeed/Megatron☆19Updated 2 weeks ago
- Scalable long-context LLM decoding that leverages sparsity—by treating the KV cache as a vector storage system.☆77Updated last week
- ☆85Updated 3 years ago
- Official Repo for "SplitQuant / LLM-PQ: Resource-Efficient LLM Offline Serving on Heterogeneous GPUs via Phase-Aware Model Partition and …☆34Updated this week
- [DAC'25] Official implement of "HybriMoE: Hybrid CPU-GPU Scheduling and Cache Management for Efficient MoE Inference"☆68Updated 2 months ago
- An experimental parallel training platform☆54Updated last year
- A ChatGPT(GPT-3.5) & GPT-4 Workload Trace to Optimize LLM Serving Systems☆200Updated last month
- [SIGMOD 2025] PQCache: Product Quantization-based KVCache for Long Context LLM Inference☆67Updated 2 months ago
- Official repository for the paper DynaPipe: Optimizing Multi-task Training through Dynamic Pipelines☆20Updated last year
- A framework for generating realistic LLM serving workloads☆54Updated 2 months ago
- High performance Transformer implementation in C++.☆129Updated 7 months ago