FluxMoE decouples MoE expert weights from persistent GPU residency via on-demand paging, achieving up to 3x throughput gains over vLLM in memory-constrained inference without accuracy loss.
Lossless compression for LLM tensor incremental snapshots
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
citation-role summary
background 1
citation-polarity summary
years
2026 2verdicts
UNVERDICTED 2roles
background 1polarities
background 1representative citing papers
ZipCCL delivers up to 1.35x faster communication and 1.18x end-to-end speedup in LLM training through lossless compression of near-Gaussian collectives on 64-GPU clusters.
citing papers explorer
-
FluxMoE: Decoupling Expert Residency for High-Performance MoE Serving
FluxMoE decouples MoE expert weights from persistent GPU residency via on-demand paging, achieving up to 3x throughput gains over vLLM in memory-constrained inference without accuracy loss.
-
ZipCCL: Efficient Lossless Data Compression of Communication Collectives for Accelerating LLM Training
ZipCCL delivers up to 1.35x faster communication and 1.18x end-to-end speedup in LLM training through lossless compression of near-Gaussian collectives on 64-GPU clusters.