Trendoday

Tideway: Phase-Aware Tiered Memory Management for MoE Inference on Consumer Devices

Large language model (LLM) agents are rapidly becoming a dominant paradigm for AI applications. Recently, there is a growing demand to deploy large-scale Mixture-of-Experts (MoE) models locally on con

  • Upvotes4
–Rank now
#1Peak rank
2Days on list
150Score · 7d
Nov 19, 2026First seen
Rank · 7 days

About this paper

TIDEWAY enables efficient inference of large Mixture-of-Experts models on memory-constrained consumer devices by treating host RAM and GPU VRAM as a unified memory pool with phase-aware layouts and optimized CUDA Graph execution under expert paging.

Large language model (LLM) agents are rapidly becoming a dominant paradigm for AI applications. Recently, there is a growing demand to deploy large-scale Mixture-of-Experts (MoE) models locally on consumer PCs to balance data privacy, API costs, and inference efficiency. However, unlike datacenter servers with abundant DRAM, consumer devices are tightly constrained in both GPU VRAM and host RAM, forcing massive MoE weights to spill across VRAM, host RAM, and SSD. Under such spilling, we observe that existing systems fail to fully utilize this memory hierarchy, treating the tiers as independent caches and thrashing in inference. We present TIDEWAY, a system for efficient inference of large…

large language model (LLM) agentsMixture-of-Experts (MoE) modelsinferenceGPU VRAMhost RAMprefilldecodephase-aware memory layoutsCUDA Graph executionexpert paging
Upvotes
4
Authors
10
Published
Nov 10, 2026

Summary and abstract excerpt via Hugging Face Papers ↗ · arXiv.

Trending history

Tideway: Phase-Aware Tiered Memory Management for MoE Inference on Consumer Devices was first seen on Hugging Face · Daily papers on Nov 19, 2026. Since then it has appeared in 12 checks on 2 different days. Its best position was #1 of 30, reached on Nov 19, 09:29 UTC. It dropped off the list after Nov 20, 03:29 UTC.

Rank history

Position on Hugging Face · Daily papers at every check (lower is better).

#1#11#20#30 05:2810:5916:2921:5903:29 Nov 19, 05:28 UTC · #22Nov 19, 07:29 UTC · #18Nov 19, 09:29 UTC · #1Nov 19, 11:29 UTC · #17Nov 19, 13:29 UTC · #17Nov 19, 15:29 UTC · #20Nov 19, 17:29 UTC · #21Nov 19, 19:29 UTC · #22Nov 19, 21:29 UTC · #21Nov 19, 23:29 UTC · #22Nov 20, 01:29 UTC · #24Nov 20, 03:29 UTC · #24

Last 14 days

Best rank per day (UTC).

Nov 9today

Metric history

Platform metric as reported by Hugging Face.

0204060 05:2810:5916:2921:5903:29 Nov 19, 05:28 UTC · 4Nov 19, 07:29 UTC · 5Nov 19, 09:29 UTC · 60Nov 19, 11:29 UTC · 6Nov 19, 13:29 UTC · 6Nov 19, 15:29 UTC · 6Nov 19, 17:29 UTC · 7Nov 19, 19:29 UTC · 7Nov 19, 21:29 UTC · 8Nov 19, 23:29 UTC · 8Nov 20, 01:29 UTC · 8Nov 20, 03:29 UTC · 8

Questions

Is Tideway: Phase-Aware Tiered Memory Management for MoE Inference on Consumer Devices trending on HF Daily Papers right now?

Not at the moment. It was last on the list on Nov 20, 03:29 UTC, after 2 days of trending.

What is the highest rank Tideway: Phase-Aware Tiered Memory Management for MoE Inference on Consumer Devices reached?

#1 on Hugging Face · Daily papers, on Nov 19, 09:29 UTC. Trendoday checks this list every 120 minutes.

How long has Tideway: Phase-Aware Tiered Memory Management for MoE Inference on Consumer Devices been trending?

It was first seen on Nov 19, 2026 and has been on the list on 2 days (12 checks) since.