The most powerful and modular diffusion model GUI, api and backend with a graph/nodes interface. The fastest local inference engine in the world.
Adjust flux mem usage factor a bit. (#9588)
comfyanonymous4aa79dbf2c51
diff --git a/comfy/supported_models.py b/comfy/supported_models.pyindex ce571e6cb..76260de00 100644--- a/comfy/supported_models.py+++ b/comfy/supported_models.py@@ -700,7 +700,7 @@ class Flux(supported_models_base.BASE):unet_extra_config = {}latent_format = latent_formats.Flux- memory_usage_factor = 2.8+ memory_usage_factor = 3.1 # TODO: debug why flux mem usage is so weird on windows.supported_inference_dtypes = [torch.bfloat16, torch.float16, torch.float32]