ComfyUI Node
Liquid Video (paid API)
External paid API. Each run buys one video from Liquid Inference, an auction exchange. The seller that wins the auction makes the video, and your account pays for it, up to the cap. A stop in ComfyUI does not stop the job or its charge. Set the API key in Settings > Liquid Inference.
Liquid Video (paid API)
- start_image
- video
- job_id
- receipt
◄prompt►
◄model▾►
◄sizeauto►
◄secondsauto►
◄cap_usd1.00►
◄retention_policynone►
CategoryLiquid Inference
Inputs (7)
| Name | Type | Default | Description |
|---|---|---|---|
| prompt | STRING | — | |
| model | COMBO | A video model that a seller quotes now. The list comes from the exchange and refreshes each minute. | |
| size | COMBO | auto | auto lets each seller use its default size. An explicit size matches only the sellers that declared it. |
| seconds | COMBO | auto | The length of the video. auto uses the shortest length that a seller declares. The charge is for each second. |
| cap_usd | FLOAT | 1.000.01–100 | The most, in USD, that this node can spend each time it runs. Video is priced for each second, so the charge is the price of the length that you ask for. |
| start_imageopt | IMAGE | The first frame of the video. Only the models that read an image input can take it. Use an image with the shape of the size. | |
| retention_policyopt | COMBO | none | none excludes providers that can keep your content. allowed permits those providers. Select allowed only if you accept that they can keep your prompt, start image and video. |
Outputs (3)
| Name | Type | Description |
|---|---|---|
| video | VIDEO | The video file. Connect it to Save Video, or to Get Video Components for the frames. |
| job_id | STRING | The exchange's ID for this job. A Liquid Video Job node can get the video again with this ID, with no new charge. Give it to support with each question about a charge. |
| receipt | STRING | JSON with the model, the seller, the charge, the cap and the file expiry for this job. |