Skip to content

The most powerful and modular diffusion model GUI, api and backend with a graph/nodes interface. The fastest local inference engine in the world.

Fix mask issue in some attention functions.

comfyanonymous2fd9c1308a2c
diff --git a/comfy/ldm/modules/attention.py b/comfy/ldm/modules/attention.py
index 7b4ee2157..885b24019 100644
--- a/comfy/ldm/modules/attention.py
+++ b/comfy/ldm/modules/attention.py
@@ -299,7 +299,10 @@ def attention_split(q, k, v, heads, mask=None, attn_precision=None, skip_reshape
if len(mask.shape) == 2:
s1 += mask[i:end]
else:
- s1 += mask[:, i:end]
+ if mask.shape[1] == 1:
+ s1 += mask
+ else:
+ s1 += mask[:, i:end]
s2 = s1.softmax(dim=-1).to(v.dtype)
del s1
diff --git a/comfy/ldm/modules/sub_quadratic_attention.py b/comfy/ldm/modules/sub_quadratic_attention.py
index 1bc4138c3..47b8b1510 100644
--- a/comfy/ldm/modules/sub_quadratic_attention.py
+++ b/comfy/ldm/modules/sub_quadratic_attention.py
@@ -234,6 +234,8 @@ def efficient_dot_product_attention(
def get_mask_chunk(chunk_idx: int) -> Tensor:
if mask is None:
return None
+ if mask.shape[1] == 1:
+ return mask
chunk = min(query_chunk_size, q_tokens)
return mask[:,chunk_idx:chunk_idx + chunk]