ComfyUI Node: Flex Attention (torch.compile)

Authored by sorryhyun

Created

Updated

1 stars

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

attention

Inputs

model MODEL

Outputs

MODEL

Extension: Flex Attention

Flex Attention (torch.compile) node for ComfyUI. Uses PyTorch's built-in flex_attention with torch.compile — no extra dependencies, works on any GPU. Falls back to PyTorch SDPA when masks are present.

Authored by sorryhyun

Looking for a different node?

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.

Learn more