Compare commits

..

5 Commits

5 changed files with 74 additions and 154 deletions

View File

@ -174,7 +174,7 @@ class Ideogram4Transformer(nn.Module):
llm = self.llm_cond_proj(llm) * text_mask
h[:, :L_text] = h[:, :L_text] + llm
h = h + self.embed_image_indicator((indicator == OUTPUT_IMAGE_INDICATOR).to(torch.long))
h = h + self.embed_image_indicator((indicator == OUTPUT_IMAGE_INDICATOR).to(torch.long), out_dtype=h.dtype)
# Qwen3-VL interleaved MRoPE; position_ids (B, L, 3) -> (3, L) (same across batch).
freqs_cis = precompute_freqs_cis(
@ -235,7 +235,7 @@ class Ideogram4Transformer2DModel(Ideogram4Transformer):
def _run_conditional(self, x_chunk, context_chunk, attn_mask_chunk, t_chunk, gh, gw, transformer_options):
B = x_chunk.shape[0]
device = x_chunk.device
img_tokens = self._img_to_tokens(x_chunk).to(self.dtype)
img_tokens = self._img_to_tokens(x_chunk)
L_img = img_tokens.shape[1]
L_text = context_chunk.shape[1]
L = L_text + L_img
@ -268,7 +268,7 @@ class Ideogram4Transformer2DModel(Ideogram4Transformer):
def _run_image_only(self, x_chunk, t_chunk, gh, gw, transformer_options):
B = x_chunk.shape[0]
device = x_chunk.device
img_tokens = self._img_to_tokens(x_chunk).to(self.dtype)
img_tokens = self._img_to_tokens(x_chunk)
L_img = img_tokens.shape[1]
position_ids = self._image_position_ids(gh, gw, device).unsqueeze(0).expand(B, L_img, 3)

View File

@ -285,7 +285,7 @@ class AudioSaveHelper:
results = []
for batch_number, waveform in enumerate(audio["waveform"].cpu()):
filename_with_batch_num = filename.replace("%batch_num%", str(batch_number))
file = f"{filename_with_batch_num}_{counter:05}_.{format}"
file = f"{filename_with_batch_num}_{counter:05}.{format}"
output_path = os.path.join(full_output_folder, file)
# Use original sample rate initially

View File

@ -158,7 +158,7 @@ class SaveAudio(IO.ComfyNode):
return IO.Schema(
node_id="SaveAudio",
search_aliases=["export flac"],
display_name="Save Audio (FLAC)",
display_name="Save Audio (FLAC) (Deprecated)",
category="audio",
essentials_category="Audio",
inputs=[
@ -167,6 +167,7 @@ class SaveAudio(IO.ComfyNode):
],
hidden=[IO.Hidden.prompt, IO.Hidden.extra_pnginfo],
is_output_node=True,
is_deprecated=True,
)
@classmethod
@ -186,7 +187,7 @@ class SaveAudioMP3(IO.ComfyNode):
return IO.Schema(
node_id="SaveAudioMP3",
search_aliases=["export mp3"],
display_name="Save Audio (MP3)",
display_name="Save Audio (MP3) (Deprecated)",
category="audio",
essentials_category="Audio",
inputs=[
@ -196,6 +197,7 @@ class SaveAudioMP3(IO.ComfyNode):
],
hidden=[IO.Hidden.prompt, IO.Hidden.extra_pnginfo],
is_output_node=True,
is_deprecated=True,
)
@classmethod
@ -217,7 +219,7 @@ class SaveAudioOpus(IO.ComfyNode):
return IO.Schema(
node_id="SaveAudioOpus",
search_aliases=["export opus"],
display_name="Save Audio (Opus)",
display_name="Save Audio (Opus) (Deprecated)",
category="audio",
inputs=[
IO.Audio.Input("audio"),
@ -226,6 +228,7 @@ class SaveAudioOpus(IO.ComfyNode):
],
hidden=[IO.Hidden.prompt, IO.Hidden.extra_pnginfo],
is_output_node=True,
is_deprecated=True,
)
@classmethod
@ -241,6 +244,54 @@ class SaveAudioOpus(IO.ComfyNode):
save_opus = execute # TODO: remove
class SaveAudioAdvanced(IO.ComfyNode):
@classmethod
def define_schema(cls):
return IO.Schema(
node_id="SaveAudioAdvanced",
search_aliases=["save audio", "export audio", "output audio", "write audio", "flac", "mp3", "opus"],
display_name="Save Audio (Advanced)",
description="Saves the input audio to your ComfyUI output directory.",
category="audio",
inputs=[
IO.Audio.Input("audio", tooltip="The audio to save."),
IO.String.Input(
"filename_prefix",
default="audio/ComfyUI",
tooltip=(
"The prefix for the file to save. May include formatting tokens "
"such as %date:yyyy-MM-dd%."
),
),
IO.DynamicCombo.Input(
"format",
options=[
IO.DynamicCombo.Option("flac", []),
IO.DynamicCombo.Option("mp3", [
IO.Combo.Input("quality", options=["V0", "128k", "320k"], default="V0"),
]),
IO.DynamicCombo.Option("opus", [
IO.Combo.Input("quality", options=["64k", "96k", "128k", "192k", "320k"], default="128k"),
]),
],
tooltip="The file format in which to save the audio.",
),
],
hidden=[IO.Hidden.prompt, IO.Hidden.extra_pnginfo],
is_output_node=True,
)
@classmethod
def execute(cls, audio, filename_prefix: str, format: dict) -> IO.NodeOutput:
file_format = format.get("format", None)
quality = format.get("quality", None)
if quality:
ui=UI.AudioSaveHelper.get_save_audio_ui(audio, filename_prefix=filename_prefix, cls=cls, format=file_format, quality=quality)
else:
ui=UI.AudioSaveHelper.get_save_audio_ui(audio, filename_prefix=filename_prefix, cls=cls, format=file_format)
return IO.NodeOutput(ui=ui)
class PreviewAudio(IO.ComfyNode):
@classmethod
def define_schema(cls):
@ -822,6 +873,7 @@ class AudioExtension(ComfyExtension):
SaveAudio,
SaveAudioMP3,
SaveAudioOpus,
SaveAudioAdvanced,
LoadAudio,
PreviewAudio,
ConditioningStableAudio,

View File

@ -3,11 +3,6 @@ components:
Asset:
description: Represents a user-owned asset (image, video, or other generated output).
properties:
asset_hash:
deprecated: true
description: 'Deprecated: use hash instead. Blake3 hash of the asset content.'
pattern: ^blake3:[a-f0-9]{64}$
type: string
created_at:
description: Timestamp when the asset was created
format: date-time
@ -17,7 +12,7 @@ components:
nullable: true
type: string
hash:
description: Blake3 hash of the asset content. Preferred over asset_hash.
description: Blake3 hash of the asset content.
pattern: ^blake3:[a-f0-9]{64}$
type: string
id:
@ -139,17 +134,12 @@ components:
AssetUpdated:
description: Response returned when an existing asset is successfully updated.
properties:
asset_hash:
deprecated: true
description: 'Deprecated: use hash instead. Blake3 hash of the asset content.'
pattern: ^blake3:[a-f0-9]{64}$
type: string
display_name:
description: Display name of the asset. Mirrors name for backwards compatibility.
nullable: true
type: string
hash:
description: Blake3 hash of the asset content. Preferred over asset_hash.
description: Blake3 hash of the asset content.
pattern: ^blake3:[a-f0-9]{64}$
type: string
id:
@ -1518,17 +1508,11 @@ paths:
schema:
default: true
type: boolean
- description: Filter assets by exact content hash. Preferred over asset_hash.
- description: Filter assets by exact content hash.
in: query
name: hash
schema:
type: string
- deprecated: true
description: 'Deprecated: use hash instead. Filter assets by exact content hash.'
in: query
name: asset_hash
schema:
type: string
- description: |
Opaque cursor for keyset pagination. Pass the `next_cursor` value
from the previous response to fetch the next page. When provided,
@ -1571,42 +1555,12 @@ paths:
- file
post:
description: |
Uploads a new asset to the system with associated metadata.
Supports two upload methods:
1. Direct file upload (multipart/form-data)
2. URL-based upload (application/json with source: "url")
Creates a new asset from a direct file upload (multipart/form-data) with associated metadata.
If an asset with the same hash already exists, returns the existing asset.
operationId: uploadAsset
operationId: createAsset
requestBody:
content:
application/json:
schema:
properties:
name:
description: Display name for the asset (used to determine file extension)
type: string
preview_id:
description: Optional preview asset ID
format: uuid
type: string
tags:
description: Freeform tags for the asset. Common types include "models", "input", "output", and "temp", but any tag can be used in any order.
items:
type: string
type: array
url:
description: HTTP/HTTPS URL to download the asset from
format: uri
type: string
user_metadata:
additionalProperties: true
description: Custom metadata to store with the asset
type: object
required:
- url
- name
type: object
multipart/form-data:
schema:
properties:
@ -1614,6 +1568,10 @@ paths:
description: The asset file to upload
format: binary
type: string
hash:
description: Content hash of the file.
pattern: ^(blake3|sha256):[a-f0-9]{64}$
type: string
id:
description: Optional asset ID for idempotent creation. If provided and asset exists, returns existing asset.
format: uuid
@ -1629,10 +1587,8 @@ paths:
format: uuid
type: string
tags:
description: Freeform tags for the asset. Common types include "models", "input", "output", and "temp", but any tag can be used in any order.
items:
type: string
type: array
description: JSON-encoded array of freeform tag strings, e.g. '["models","checkpoint"]'. Common types include "models", "input", "output", and "temp", but any tag can be used in any order.
type: string
user_metadata:
description: Custom JSON metadata as a string
type: string
@ -1652,25 +1608,13 @@ paths:
application/json:
schema:
$ref: '#/components/schemas/ErrorResponse'
description: Invalid request (bad file, invalid URL, invalid content type, etc.)
description: Invalid request (bad file, invalid content type, etc.)
"401":
content:
application/json:
schema:
$ref: '#/components/schemas/ErrorResponse'
description: Unauthorized
"403":
content:
application/json:
schema:
$ref: '#/components/schemas/ErrorResponse'
description: Source URL requires authentication or access denied
"404":
content:
application/json:
schema:
$ref: '#/components/schemas/ErrorResponse'
description: Source URL not found
"413":
content:
application/json:
@ -1683,19 +1627,13 @@ paths:
schema:
$ref: '#/components/schemas/ErrorResponse'
description: Unsupported media type
"422":
content:
application/json:
schema:
$ref: '#/components/schemas/ErrorResponse'
description: Download failed due to network error or timeout
"500":
content:
application/json:
schema:
$ref: '#/components/schemas/ErrorResponse'
description: Internal server error
summary: Upload a new asset
summary: Create a new asset
tags:
- file
/api/assets/{id}:
@ -1783,7 +1721,7 @@ paths:
description: |
Updates an asset's metadata. At least one field must be provided.
Only name, mime_type, preview_id, and user_metadata can be updated.
For tag management, use the dedicated PUT /api/assets/{id}/tags endpoint.
For tag management, use POST (add) and DELETE (remove) /api/assets/{id}/tags.
operationId: updateAsset
parameters:
- description: Asset ID
@ -1982,76 +1920,6 @@ paths:
summary: Add tags to asset
tags:
- file
put:
description: Adds and removes tags from an asset in a single operation
operationId: updateAssetTags
parameters:
- description: Asset ID
in: path
name: id
required: true
schema:
format: uuid
type: string
requestBody:
content:
application/json:
schema:
description: At least one of add or remove must contain items. Empty arrays are allowed when the other array has items.
minProperties: 1
properties:
add:
description: Tags to add to the asset. Can be empty if remove has items.
items:
type: string
type: array
remove:
description: Tags to remove from the asset. Can be empty if add has items.
items:
type: string
type: array
type: object
required: true
responses:
"200":
content:
application/json:
schema:
$ref: '#/components/schemas/TagsModificationResponse'
description: Tags updated successfully
"400":
content:
application/json:
schema:
$ref: '#/components/schemas/ErrorResponse'
description: Invalid request
"401":
content:
application/json:
schema:
$ref: '#/components/schemas/ErrorResponse'
description: Unauthorized
"404":
content:
application/json:
schema:
$ref: '#/components/schemas/ErrorResponse'
description: Asset not found
"422":
content:
application/json:
schema:
$ref: '#/components/schemas/ErrorResponse'
description: Reserved tag validation error
"500":
content:
application/json:
schema:
$ref: '#/components/schemas/ErrorResponse'
description: Internal server error
summary: Update asset tags
tags:
- file
/api/assets/from-hash:
post:
description: |

View File

@ -1,4 +1,4 @@
comfyui-frontend-package==1.44.19
comfyui-frontend-package==1.45.15
comfyui-workflow-templates==0.9.98
comfyui-embedded-docs==0.5.2
torch