method · first seen 4 Oct 2026
pipeline parallelism
Also: PP, GPipe, micro-batch pipelining
Splitting a computation into stages across devices that process different inputs concurrently.
Mentions per week
PapersCodeDiscussion
Earliest sources
- Characterizing Parallelism Strategies in LLM Inference: Fundamental Compute-Communication Trade-offsPapers · 4 Oct 2026 · Javad Mirzaei, Jeebak Mitra
- joshuaswarren/omarchy-clusterCode · 6 Oct 2026 · joshuaswarren · 25 stars
- AgentWeave: Efficient Distributed Agent Serving via Flow DecompositionPapers · 10 Oct 2026 · Kaibin Guo, Pengtu Li, Zicong Hong, Zhiyuan Fang, Wuhui Chen, Rachid Guerraoui,…
- BSIPipe: Bidirectional Sequence-Level Interleaved Pipeline Parallelism for Long-Sequence Large Language Model TrainingPapers · 10 Oct 2026 · Yuzhou Ni, Ling Chen
- Scaling LLM Inference via Parallelism: Throughput-Latency Trade-offPapers · 14 Oct 2026 · Javad Mirzaei, Jeebak Mitra
- Exact Makespan Bounds and the GPipe Homogeneity Bias for Heterogeneous LLM Inference PipelinesPapers · 15 Oct 2026 · Zhu Wenbo
Latest sources
- Benchmarking LLM Inference on Desktop-Scale NVIDIA Grace Blackwell ClustersPapers · 15 Nov 2026 · Kevin Bhimani, Christian Escobar, Daniel Mihovch, Rudolf Eigenmann, Sunita Chand…
- ProprietaryLegal/Volta-sPLIt-DeepSeek-V4.1-Flash-1017-8xV100-SXM2-PP2xTP4Code · 3 Nov 2026 · ProprietaryLegal · 33 stars
- GeoTrain: Energy-Aware Automatic Parallelism for Foundation Model Training on Heterogeneous Geo-Distributed ResourcesPapers · 19 Oct 2026 · Bui-Khanh Hoang, TianZhang He, Mohammad Goudarzi
- ETH: Data-Balanced Heterogeneous Pipeline Parallelism for Training Large ModelsPapers · 15 Oct 2026 · Mianjie Yu, Zhongmin Zhao, Zizhao Mo, Jianxiong Liao, Wenxuan Li, Yulin Qiao, Hu…
- Exact Makespan Bounds and the GPipe Homogeneity Bias for Heterogeneous LLM Inference PipelinesPapers · 15 Oct 2026 · Zhu Wenbo
- Scaling LLM Inference via Parallelism: Throughput-Latency Trade-offPapers · 14 Oct 2026 · Javad Mirzaei, Jeebak Mitra
- BSIPipe: Bidirectional Sequence-Level Interleaved Pipeline Parallelism for Long-Sequence Large Language Model TrainingPapers · 10 Oct 2026 · Yuzhou Ni, Ling Chen
- AgentWeave: Efficient Distributed Agent Serving via Flow DecompositionPapers · 10 Oct 2026 · Kaibin Guo, Pengtu Li, Zicong Hong, Zhiyuan Fang, Wuhui Chen, Rachid Guerraoui,…
- joshuaswarren/omarchy-clusterCode · 6 Oct 2026 · joshuaswarren · 25 stars
- Characterizing Parallelism Strategies in LLM Inference: Fundamental Compute-Communication Trade-offsPapers · 4 Oct 2026 · Javad Mirzaei, Jeebak Mitra