OmniHuman 1.5 (BudgetPixel):
BudgetPixel_omni-human-1.5 is a video-generation node for a workflow that combines visual and sound material. It prepares your settings for a remote generation service, submits a job, waits for the result, and makes the generated video available in your workflow. This lets you build a video-creation step alongside other nodes without handling the service request and job polling yourself. The provided source does not specify the model’s animation style, duration, synchronization behavior, or processing limits, so those characteristics should be judged from your results rather than assumed.
OmniHuman 1.5 (BudgetPixel) Input Parameters:
image
A required IMAGE connection that supplies the visual material for the generation request. Connect an image-producing node before running this node; if no image is connected, the request cannot be prepared. The provided interface does not specify an image-size limit, a default image, or whether multiple images are accepted through this socket.
audio
A required AUDIO connection that supplies sound material for the generation request. Connect an audio-producing or audio-loading node before running this node; if it is missing, the request cannot be prepared. The provided interface does not specify a duration limit, a default audio clip, or how the service uses the sound in the finished video.
output_resolution
An optional integer-choice control for requesting an output resolution. Choose from the options displayed in the node to specify a preference for the generated video. The supplied source does not include the actual choices, their dimensions, or a default, so no minimum, maximum, or recommended value can be stated reliably.
prompt
An optional text field for instructions sent with the generation request. Use it to express the result you want, while checking the finished video to see how the service interprets your wording. When this field is empty, the node omits it from the request. No default text, length limit, or prompt syntax is specified in the provided source.
force_rerun
An optional Boolean control with a default of False. Turn it on when you want the node to be treated as changed on each evaluation, causing a fresh execution even when the other inputs have not changed. Leave it off when you do not need that behavior. Its options are True and False; it does not itself specify how the service generates the video.
OmniHuman 1.5 (BudgetPixel) Output Parameters:
video
A VIDEO output containing the result returned after the submitted job completes and the node extracts its video result. Connect it to a compatible preview, processing, or saving node to inspect or use the generated video. The provided source does not specify the video’s file format, duration, frame rate, or dimensions.
OmniHuman 1.5 (BudgetPixel) Usage Tips:
- Connect both
imageandaudiobefore running the node. Neither is optional, and leaving either disconnected prevents the request from being prepared. - Start with a concise
promptthat describes your intended result, then revise it based on the video you receive. An empty optional prompt is omitted from the request. - Check the choices shown for
output_resolutionin your installed node. The supplied source does not establish which resolutions are available. - Keep
force_rerunoff during ordinary workflow evaluation. Enable it when you deliberately want a new execution with otherwise unchanged settings. - Allow time for the remote job to finish. The node submits the request and polls for completion, but the provided source does not specify a fixed completion time.
OmniHuman 1.5 (BudgetPixel) Common Errors and Solutions:
Input 'image' is required — connect a IMAGE.
- Explanation: The node’s required visual input was not connected. The exact socket wording in the message depends on the node’s media mapping, which is not included in the provided source.
- Solution: Connect an
IMAGEoutput toimageand run the node again.
Input 'audio' is required — connect a AUDIO.
- Explanation: The node’s required sound input was not connected. The exact socket wording in the message depends on the node’s media mapping, which is not included in the provided source.
- Solution: Connect an
AUDIOoutput toaudioand run the node again.
Media preparation failed
- Explanation: The node converts connected media into values suitable for the remote request. The provided source does not expose the conversion functions or their exact error messages, so this is a possible failure category rather than a quoted message.
- Solution: Confirm that the connected nodes produce valid
IMAGEandAUDIOoutputs, then retry with media that loads successfully elsewhere in your workflow.
Remote job failed or timed out
- Explanation: The node submits a remote job and waits for it to finish. A service failure or wait timeout could stop the node before it returns a video; the provided source does not give the exact message or timeout duration.
- Solution: Read the full error shown by your workflow, check that the required media is connected, and retry. If the problem persists, check the service’s status and your access to it.
No video result was returned
- Explanation: The node relies on a result-extraction function after the remote job responds. Its implementation is not provided, so the precise conditions and error text for a missing result are unknown.
- Solution: Inspect the full job error, if one is shown, and retry with valid connected media. If repeated completed jobs produce no video, report the job details to the node or service provider.
