DataStates / datastates-llmLinks
LLM checkpointing for DeepSpeed/Megatron
☆23Updated 2 months ago
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
- ☆83Updated 3 months ago
- SpotServe: Serving Generative Large Language Models on Preemptible Instances☆134Updated last year
- Microsoft Collective Communication Library☆66Updated last year
- Dynamic resources changes for multi-dimensional parallelism training☆30Updated 5 months ago
- Stateful LLM Serving☆95Updated 10 months ago
- FlexFlow Serve: Low-Latency, High-Performance LLM Serving☆71Updated 4 months ago
- [OSDI'24] Serving LLM-based Applications Efficiently with Semantic Variable☆208Updated last year
- [ICML 2024] Serving LLMs on heterogeneous decentralized clusters.☆34Updated last year
- ☆50Updated 9 months ago
- A framework for generating realistic LLM serving workloads☆98Updated 3 months ago
- ☆48Updated last year
- ☆26Updated 2 years ago
- Scalable long-context LLM decoding that leverages sparsity—by treating the KV cache as a vector storage system.☆119Updated last month
- ☆130Updated last year
- ☆76Updated last week
- Official repository for the paper DynaPipe: Optimizing Multi-task Training through Dynamic Pipelines☆19Updated 2 years ago
- Chimera: bidirectional pipeline parallelism for efficiently training large-scale models.☆70Updated 10 months ago
- AI model training on heterogeneous, geo-distributed resources☆33Updated 2 months ago
- An interference-aware scheduler for fine-grained GPU sharing☆159Updated 2 months ago
- NEO is a LLM inference engine built to save the GPU memory crisis by CPU offloading☆79Updated 7 months ago
- High-performance distributed data shuffling (all-to-all) library for MoE training and inference☆109Updated last month
- PyTorch library for cost-effective, fast and easy serving of MoE models.☆278Updated this week
- nnScaler: Compiling DNN models for Parallel Training☆124Updated 4 months ago
- [NeurIPS 2024] Efficient LLM Scheduling by Learning to Rank☆67Updated last year
- Thunder Research Group's Collective Communication Library☆47Updated 6 months ago
- Accepted to MLSys 2026☆70Updated this week
- ☆76Updated last year
- Code for MLSys 2024 Paper "SiDA-MoE: Sparsity-Inspired Data-Aware Serving for Efficient and Scalable Large Mixture-of-Experts Models"☆22Updated last year
- An experimental parallel training platform☆56Updated last year