Smart Resident Scheduler (LMSL)

Future-aware smart resident scheduler for ComfyUI low VRAM (--lowvram / --enable-dynamic-vram). Reduces model swap thrashing on 12GB GPUs such as RTX 3060.

By mxh678988·Created 23 days ago·Updated 21 days ago· 1

mxh678988/lmsl-smart-offload

Nodes3

On cloudLocal install

CategoryLMSL/模型选型, lmsl-smart-offload

Stars1

Updated21 days ago

Nodes (3)

Smart Model Advisor (LMSL / TDS SDK)

A Second Opinion on the Checkpoint You Just Picked

LMSL/模型选型

Smart Model Recommender (LMSL / TDS SDK)

Ask 'What Should I Run for Wan 2.1 Text-to-Video?'

LMSL/模型选型

Smart Resident Scheduler (LMSL)

Stop ComfyUI Evicting the Model It's About to Need

lmsl-smart-offload

Readme

Smart Resident Scheduler (LMSL)

Future-aware smart resident scheduler for ComfyUI low VRAM mode.

Reduces model swap thrashing on 12GB GPUs (e.g. RTX 3060) when running with --lowvram / --enable-dynamic-vram.

Features

Reserves resident models in VRAM to avoid reload thrashing
Future-aware scheduling based on queue preview
Lightweight: no extra dependencies beyond ComfyUI itself
TDS SDK selection bridge (v0.2.0): model health audit, intent-based recommendation nodes, and read-only HTTP endpoints under /lmsl_sdk/* (status / recommend / advise). All bridge paths fail soft and never block startup or workflow execution.

Installation

Install from ComfyUI Registry (recommended):

comfy node install lmsl-smart-offload

Or install from source:

git clone https://github.com/mxh678988/lmsl-smart-offload.git
custom_nodes/lmsl-smart-offload

Then restart ComfyUI.

Usage

The node appears in the node list after installation. Connect it into your workflow to enable future-aware resident scheduling under low VRAM mode.

License

MIT