ComfyUI Node: DGLS Model Loader
Run ComfyUI workflows without the setup
No installs, no CUDA version roulette, no GPU sitting idle on your bill. Bring a workflow and run it in the browser.
Category
loaders
Inputs
model_name
model_type
- default
- hunyuan
- unet
cast_dtype
- default
- disabled
- fp32
- fp16
- bf16
- fp8_e4m3fn
- fp8_e4m3fn_fast
- fp8_e5m2
full_cast
- disabled
- fp32
- fp16
- bf16
- fp8_e4m3
- fp8_e5m2
- nf4
- fp4
clear_model_cache BOOLEAN
verbose BOOLEAN
nuke_all_caches BOOLEAN
Outputs
MODEL
LAYERS
Extension: ComfyUI - DGLS (Dynamic GPU Layer Swapping)
Smart dynamic layer swapping between GPU and CPU for optimal inference performance with comprehensive mixed precision handling and copy-compute overlap optimization. Enables running much larger models on limited VRAM setups.
Authored by obisin
Looking for a different node?
More nodes in ComfyUI - DGLS (Dynamic GPU Layer Swapping)
Other loaders nodes
- Additional Features With Attention
- Anima EasyControl (KSampler)
- Apply Anima ControlNet-LLLite
- Apply Trained LoRA
- Requires Asset
- Load Attention couple
- Auto Model Finder (Experimental)
- Apply CDTuner
- Checkpoint File Loader (Soze)
- Checkpoint Loader Iterator
- Load Checkpoint Mix With Diffusers
- Load FP4 or NF4 Quantized Checkpoint Model
Run ComfyUI workflows without the setup
No installs, no CUDA version roulette, no GPU sitting idle on your bill. Bring a workflow and run it in the browser.