DataStates / datastates-llmLinks
LLM checkpointing for DeepSpeed/Megatron
☆21Updated last month
Alternatives and similar repositories for datastates-llm
Users that are interested in datastates-llm are comparing it to the libraries listed below
Sorting:
- A resilient distributed training framework☆96Updated last year
- ☆79Updated last month
- [ICML 2024] Serving LLMs on heterogeneous decentralized clusters.☆31Updated last year
- [OSDI'24] Serving LLM-based Applications Efficiently with Semantic Variable☆196Updated last year
- Stateful LLM Serving☆88Updated 8 months ago
- ☆43Updated 6 months ago
- Code for MLSys 2024 Paper "SiDA-MoE: Sparsity-Inspired Data-Aware Serving for Efficient and Scalable Large Mixture-of-Experts Models"☆21Updated last year
- NEO is a LLM inference engine built to save the GPU memory crisis by CPU offloading☆69Updated 5 months ago
- Chimera: bidirectional pipeline parallelism for efficiently training large-scale models.☆68Updated 8 months ago
- ☆48Updated last year
- Microsoft Collective Communication Library☆66Updated last year
- A framework for generating realistic LLM serving workloads☆79Updated last month
- [NeurIPS 2024] Efficient LLM Scheduling by Learning to Rank☆64Updated last year
- SpotServe: Serving Generative Large Language Models on Preemptible Instances☆132Updated last year
- Official repository for the paper DynaPipe: Optimizing Multi-task Training through Dynamic Pipelines☆20Updated last year
- A lightweight design for computation-communication overlap.☆187Updated last month
- PyTorch library for cost-effective, fast and easy serving of MoE models.☆259Updated last month
- An interference-aware scheduler for fine-grained GPU sharing☆152Updated 9 months ago
- Dynamic resources changes for multi-dimensional parallelism training☆29Updated 3 months ago
- ☆124Updated last year
- FlexFlow Serve: Low-Latency, High-Performance LLM Serving☆63Updated 2 months ago
- PyTorch implementation of paper "Response Length Perception and Sequence Scheduling: An LLM-Empowered LLM Inference Pipeline".☆93Updated 2 years ago
- ☆151Updated 4 months ago
- ☆25Updated 2 years ago
- torchcomms: a modern PyTorch communications API☆291Updated this week
- ☆21Updated last year
- Thunder Research Group's Collective Communication Library☆42Updated 4 months ago
- Efficient Compute-Communication Overlap for Distributed LLM Inference☆62Updated 3 weeks ago
- ☆27Updated 8 months ago
- ☆140Updated last year