ComfyUI Node: FP8 / MXFP8 Model Patch (SM89+)
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
optimization
Inputs
model MODEL
mode
- fp8
- mxfp8
accumulator
- fp16
- fp32
Outputs
MODEL
Extension: ComfyUI-fp8_cublas
ComfyUI custom node that patches loaded models (Flux, SDXL, etc.) to run Linear layers in FP8 or MXFP8, unlocking tensor-core throughput on Ada Lovelace (RTX 4090) and Blackwell (RTX 5090) GPUs.
Authored by marduk191
Looking for a different node?
Other optimization nodes
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.