deepseek-ai / EPLBLinks
Expert Parallelism Load Balancer
☆1,209Updated 2 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.☆1,059Updated 3 months ago
- A bidirectional pipeline parallelism algorithm for computation-communication overlap in V3/R1 training.☆2,807Updated 3 months ago
- DeepGEMM: clean and efficient FP8 GEMM kernels with fine-grained scaling☆5,450Updated last week
- DeepEP: an efficient expert-parallel communication library☆7,911Updated this week
- A fast communication-overlapping library for tensor/expert parallelism on GPUs.☆986Updated 3 weeks ago
- Mooncake is the serving platform for Kimi, a leading LLM service provided by Moonshot AI.☆3,436Updated this week
- Muon is Scalable for LLM Training☆1,077Updated 2 months ago
- FlashInfer: Kernel Library for LLM Serving☆3,211Updated this week
- MoBA: Mixture of Block Attention for Long-Context LLMs☆1,798Updated 2 months ago
- DeepSeekMoE: Towards Ultimate Expert Specialization in Mixture-of-Experts Language Models☆1,731Updated last year
- Redis for LLMs☆1,560Updated this week
- Nano vLLM☆1,659Updated this week
- High-performance inference framework for large language models, focusing on efficiency, flexibility, and availability.☆1,144Updated this week
- ☆789Updated last week
- Domain-specific language designed to streamline the development of high-performance GPU/CPU/Accelerators kernels☆1,292Updated this week
- Production-tested AI infrastructure tools for efficient AGI development and community-driven innovation☆7,824Updated last month
- Distributed RL System for LLM Reasoning☆1,774Updated this week
- A throughput-oriented high-performance serving framework for LLMs☆825Updated 2 weeks ago
- Materials for learning SGLang☆443Updated this week
- 🐳 Efficient Triton implementations for "Native Sparse Attention: Hardware-Aligned and Natively Trainable Sparse Attention"☆700Updated 3 months ago
- A lightweight data processing framework built on DuckDB and 3FS.☆4,697Updated 3 months ago
- The official repo of MiniMax-Text-01 and MiniMax-VL-01, large-language-model & vision-language-model based on Linear Attention☆2,891Updated this week
- Distributed Compiler Based on Triton for Parallel Systems☆829Updated this week
- Official Repo for Open-Reasoner-Zero☆1,967Updated 2 weeks ago
- A high-performance distributed file system designed to address the challenges of AI training and inference workloads.☆9,022Updated this week
- FlashMLA: Efficient MLA decoding kernels☆11,607Updated last month
- A PyTorch Native LLM Training Framework☆819Updated 5 months ago
- [NeurIPS'24 Spotlight, ICLR'25, ICML'25] To speed up Long-context LLMs' inference, approximate and dynamic sparse calculate the attention…☆1,055Updated this week
- Production ready LLM model compression/quantization toolkit with hw accelerated inference support for both cpu/gpu via HF, vLLM, and SGLa…☆625Updated this week
- Fast, Flexible and Portable Structured Generation☆1,029Updated last week