deepseek-ai / DualPipeLinks
A bidirectional pipeline parallelism algorithm for computation-communication overlap in DeepSeek V3/R1 training.
☆2,866Updated 6 months ago
Alternatives and similar repositories for DualPipe
Users that are interested in DualPipe are comparing it to the libraries listed below
Sorting:
- Expert Parallelism Load Balancer☆1,274Updated 6 months ago
- DeepGEMM: clean and efficient FP8 GEMM kernels with fine-grained scaling☆5,780Updated last week
- Analyze computation-communication overlap in V3/R1.☆1,101Updated 6 months ago
- DeepEP: an efficient expert-parallel communication library☆8,584Updated last week
- FlashMLA: Efficient Multi-head Latent Attention Kernels☆11,785Updated last week
- Production-tested AI infrastructure tools for efficient AGI development and community-driven innovation☆7,919Updated 4 months ago
- FlashInfer: Kernel Library for LLM Serving☆3,861Updated this week
- Mooncake is the serving platform for Kimi, a leading LLM service provided by Moonshot AI.☆4,054Updated this week
- Domain-specific language designed to streamline the development of high-performance GPU/CPU/Accelerators kernels☆3,197Updated this week
- A high-performance distributed file system designed to address the challenges of AI training and inference workloads.☆9,351Updated 2 weeks ago
- A fast communication-overlapping library for tensor/expert parallelism on GPUs.☆1,137Updated last month
- MoBA: Mixture of Block Attention for Long-Context LLMs☆1,912Updated 6 months ago
- DeepSeekMoE: Towards Ultimate Expert Specialization in Mixture-of-Experts Language Models☆1,804Updated last year
- A Datacenter Scale Distributed Inference Serving Framework☆5,244Updated this week
- Muon is Scalable for LLM Training☆1,321Updated 2 months ago
- A lightweight data processing framework built on DuckDB and 3FS.☆4,788Updated 7 months ago
- High-performance inference framework for large language models, focusing on efficiency, flexibility, and availability.☆1,293Updated this week
- Mirage Persistent Kernel: Compiling LLMs into a MegaKernel☆1,866Updated this week
- slime is an LLM post-training framework for RL Scaling.☆2,023Updated this week
- Lightning-Fast RL for LLM Reasoning and Agents. Made Simple & Flexible.☆2,736Updated last week
- My learning notes/codes for ML SYS.☆3,808Updated this week
- Distributed Compiler based on Triton for Parallel Systems☆1,152Updated last week
- Democratizing Reinforcement Learning for LLMs☆4,414Updated this week
- 🐳 Efficient Triton implementations for "Native Sparse Attention: Hardware-Aligned and Natively Trainable Sparse Attention"☆885Updated 6 months ago
- The official repo of MiniMax-Text-01 and MiniMax-VL-01, large-language-model & vision-language-model based on Linear Attention☆3,157Updated 3 months ago
- A PyTorch Native LLM Training Framework☆872Updated 3 weeks ago
- A library for accelerating Transformer models on NVIDIA GPUs, including using 8-bit floating point (FP8) precision on Hopper, Ada and Bla…☆2,763Updated this week
- Nano vLLM☆6,999Updated last month
- Transformers-compatible library for applying various compression algorithms to LLMs for optimized deployment with vLLM☆2,042Updated this week
- Fast, Flexible and Portable Structured Generation☆1,288Updated last week