pith. sign in

Xpipe: Efficient pipeline model parallelism for multi-gpu dnn training

2 Pith papers cite this work. Polarity classification is still indexing.

2 Pith papers citing it

citation-role summary

background 1

citation-polarity summary

fields

cs.DC 1 cs.LG 1

years

2026 2

roles

background 1

polarities

background 1

clear filters

representative citing papers

Efficient Training on Multiple Consumer GPUs with RoundPipe

cs.DC · 2026-04-29 · conditional · novelty 8.0

RoundPipe achieves near-zero-bubble pipeline parallelism for LLM training on consumer GPUs by dynamically dispatching computation stages round-robin, yielding 1.48-2.16x speedups and enabling 235B model fine-tuning on 8x RTX 4090.

citing papers explorer

Showing 1 of 1 citing paper after filters.

  • Efficient Training on Multiple Consumer GPUs with RoundPipe cs.DC · 2026-04-29 · conditional · none · ref 16

    RoundPipe achieves near-zero-bubble pipeline parallelism for LLM training on consumer GPUs by dynamically dispatching computation stages round-robin, yielding 1.48-2.16x speedups and enabling 235B model fine-tuning on 8x RTX 4090.