Merge fa87f263ce into ff838657fa

comfy/ldm/lightricks/model.py

@@ -456,9 +456,8 @@ class LTXVModel(torch.nn.Module):
         x = self.patchify_proj(x)
         timestep = timestep * 1000.0
 
-        attention_mask = 1.0 - attention_mask.to(x.dtype).reshape((attention_mask.shape[0], 1, -1, attention_mask.shape[-1]))
-        attention_mask = attention_mask.masked_fill(attention_mask.to(torch.bool), float("-inf"))  # not sure about this
-        # attention_mask = (context != 0).any(dim=2).to(dtype=x.dtype)
+        if attention_mask is not None and not torch.is_floating_point(attention_mask):
+            attention_mask = (attention_mask - 1).to(x.dtype).reshape((attention_mask.shape[0], 1, -1, attention_mask.shape[-1])) * torch.finfo(x.dtype).max
 
         pe = precompute_freqs_cis(indices_grid, dim=self.inner_dim, out_dtype=x.dtype)
 

comfy/sample.py

@@ -5,12 +5,15 @@ import comfy.utils
 import numpy as np
 import logging
 
-def prepare_noise(latent_image, seed, noise_inds=None):
+def prepare_noise(latent_image, seed, noise_inds=None, disable_noise=False):
     """
     creates random noise given a latent image and a seed.
     optional arg skip can be used to skip and discard x number of noise generations for a given seed
     """
     generator = torch.manual_seed(seed)
+    if disable_noise:
+        return torch.zeros(latent_image.size(), dtype=latent_image.dtype, layout=latent_image.layout, device="cpu")
+
     if noise_inds is None:
         return torch.randn(latent_image.size(), dtype=latent_image.dtype, layout=latent_image.layout, generator=generator, device="cpu")
 

comfy_extras/nodes_custom_sampler.py

@@ -418,7 +418,7 @@ class Noise_EmptyNoise:
 
     def generate_noise(self, input_latent):
         latent_image = input_latent["samples"]
-        return torch.zeros(latent_image.shape, dtype=latent_image.dtype, layout=latent_image.layout, device="cpu")
+        return comfy.sample.prepare_noise(latent_image, self.seed, disable_noise=True)
 
 
 class Noise_RandomNoise:

nodes.py

@@ -1485,11 +1485,8 @@ def common_ksampler(model, seed, steps, cfg, sampler_name, scheduler, positive,
     latent_image = latent["samples"]
     latent_image = comfy.sample.fix_empty_latent_channels(model, latent_image)
 
-    if disable_noise:
-        noise = torch.zeros(latent_image.size(), dtype=latent_image.dtype, layout=latent_image.layout, device="cpu")
-    else:
-        batch_inds = latent["batch_index"] if "batch_index" in latent else None
-        noise = comfy.sample.prepare_noise(latent_image, seed, batch_inds)
+    batch_inds = latent["batch_index"] if "batch_index" in latent else None
+    noise = comfy.sample.prepare_noise(latent_image, seed, batch_inds, disable_noise)
 
     noise_mask = None
     if "noise_mask" in latent: