deepseek-ai / EPLBLinks
Expert Parallelism Load Balancer
☆1,240Updated 4 months ago
Alternatives and similar repositories for EPLB
Users that are interested in EPLB are comparing it to the libraries listed below
Sorting:
- Analyze computation-communication overlap in V3/R1.☆1,081Updated 4 months ago
- A bidirectional pipeline parallelism algorithm for computation-communication overlap in V3/R1 training.☆2,835Updated 4 months ago
- DeepGEMM: clean and efficient FP8 GEMM kernels with fine-grained scaling☆5,562Updated this week
- A fast communication-overlapping library for tensor/expert parallelism on GPUs.☆1,046Updated this week
- DeepEP: an efficient expert-parallel communication library☆8,316Updated last week
- Mooncake is the serving platform for Kimi, a leading LLM service provided by Moonshot AI.☆3,665Updated this week
- FlashInfer: Kernel Library for LLM Serving☆3,448Updated this week
- A PyTorch Native LLM Training Framework☆839Updated 3 weeks ago
- Distributed Compiler based on Triton for Parallel Systems☆930Updated this week
- Materials for learning SGLang☆515Updated 2 weeks ago
- Domain-specific language designed to streamline the development of high-performance GPU/CPU/Accelerators kernels☆1,472Updated this week
- Muon is Scalable for LLM Training☆1,223Updated 4 months ago
- Disaggregated serving system for Large Language Models (LLMs).☆654Updated 3 months ago
- MoBA: Mixture of Block Attention for Long-Context LLMs☆1,846Updated 4 months ago
- Mirage: Automatically Generating Fast GPU Kernels without Programming in Triton/CUDA☆1,629Updated this week
- A throughput-oriented high-performance serving framework for LLMs☆856Updated 3 weeks ago
- High-performance inference framework for large language models, focusing on efficiency, flexibility, and availability.☆1,173Updated last week
- slime is a LLM post-training framework aiming for RL Scaling.☆975Updated this week
- A Datacenter Scale Distributed Inference Serving Framework☆4,573Updated this week
- Official Implementation of EAGLE-1 (ICML'24), EAGLE-2 (EMNLP'24), and EAGLE-3.☆1,439Updated this week
- RTP-LLM: Alibaba's high-performance LLM inference engine for diverse applications.☆823Updated this week
- Efficient and easy multi-instance LLM serving☆454Updated this week
- 🐳 Efficient Triton implementations for "Native Sparse Attention: Hardware-Aligned and Natively Trainable Sparse Attention"☆731Updated 4 months ago
- Ring attention implementation with flash attention☆828Updated last week
- DeepSeekMoE: Towards Ultimate Expert Specialization in Mixture-of-Experts Language Models☆1,758Updated last year
- [NeurIPS'24 Spotlight, ICLR'25, ICML'25] To speed up Long-context LLMs' inference, approximate and dynamic sparse calculate the attention…☆1,080Updated this week
- NVIDIA Inference Xfer Library (NIXL)☆491Updated this week
- Perplexity GPU Kernels☆413Updated 2 weeks ago
- [MLSys'25] QServe: W4A8KV4 Quantization and System Co-design for Efficient LLM Serving; [MLSys'25] LServe: Efficient Long-sequence LLM Se…☆730Updated 4 months ago
- FlagGems is an operator library for large language models implemented in the Triton Language.☆635Updated this week