Buy images and videos in ComfyUI with Liquid Inference, with a USD cap on each run. Use Nano Banana (Gemini 3.1 Flash Image), Seedream, Seedance, Kling, Veo and FLUX when sellers offer them.

This package adds three nodes to ComfyUI:

Liquid Image (paid API) buys images.
Liquid Video (paid API) buys a video.
Liquid Video Job gets the video of a job again, with no new charge.

The paid nodes send a text prompt to Liquid Inference, an auction exchange for inference. Sellers post offers for image and video models. The exchange gives your request to the best offer that is under your price cap, and you pay for the images or the video that you get.

[!IMPORTANT] Each run of a paid node is a paid request to an external service. Your account pays for it. The prompt goes to the seller that wins the auction.

Requirements

ComfyUI 0.3.45 or newer.
A Liquid Inference API key with the spend scope, and a balance on the account.

Install

In ComfyUI, open Manager, then Custom Nodes Manager.
Search for "Liquid Inference".
Select Install, then restart ComfyUI.

To install by hand, clone this repository into ComfyUI/custom_nodes/, then restart ComfyUI. The package has no Python dependencies to install.

Sign in

For ComfyUI on this machine, open Settings > Liquid Inference, then select Sign in. Approve spend access in the new page. Select a spend limit that covers your planned runs. The key expires after seven days. When it expires, select Sign in again. Sign-in uses the default Liquid router and an HTTP loopback address.

For a remote ComfyUI server, create a key at https://inference.ai.exchange/keys. Paste that key with the steps below. Add funds at https://inference.ai.exchange/credits if your balance is too low.

Set the API key

Open Settings, then Liquid Inference.
Paste your key into API key.
Select Save.

The key goes to the ComfyUI server. The server keeps it in a file in the ComfyUI user directory, user/__liquid_inference/api_key. On Linux and macOS, only your system user can read this file. ComfyUI does not serve folders whose names start with __.

As an alternative, set the LIQUID_API_KEY environment variable before you start ComfyUI. The environment variable has priority over the stored key.

The nodes have no input for the key. ComfyUI writes each input value into the workflow file and into the metadata of each saved image. A key in an input would thus go to each person who gets one of your images or workflows.

Buy images

Double-click the canvas and search for "Liquid Image".
Write a prompt, and select a model.
Connect images to Save Image or to a different image node.
Queue the workflow.

The example_workflows folder has five workflows: text to image, text to image with an upscale, two models side by side, text to video and image to video. In ComfyUI, open them from Templates.

Inputs

| Input | Description | | --- | --- | | prompt | The text that describes the image. | | model | An image model that a seller offers now. The node gets the list from the exchange and refreshes it each minute. | | size | auto, or a size in pixels. auto lets each seller use its default size, so more offers can take the request. An explicit size goes only to the sellers that declared it. | | n | The number of images, from 1 to 4. | | cap_usd | The most that one run of the node can cost, in USD. |

The cap applies to each request, not to the workflow. A workflow with three Liquid nodes and a cap of 0.05 can cost up to 0.15 USD for each run. If you queue the workflow ten times, each run is a new request with its own cap.

ComfyUI does not run a node again when its inputs did not change. Thus a second queue of the same workflow does not buy the same images again. Change an input to buy new images.

Outputs

| Output | Description | | --- | --- | | images | The images, as a ComfyUI IMAGE batch. | | job_id | The exchange's ID for the job. Give this ID to support with each question about a charge. | | receipt | JSON with the job ID, the model, the seller, the charge in USD, your cap and the node version. |

Buy a video

Double-click the canvas and search for "Liquid Video".
Write a prompt, and select a model.
Connect video to Save Video.
Queue the workflow.

The node buys the video with one request. Then it reads the status of the job every 5 seconds, and it shows the status on the node. When the job is complete, the node gets the file from the exchange. The node waits for 30 minutes at most.

Inputs

| Input | Description | | --- | --- | | prompt | The text that describes the video. | | model | A video model that a seller offers now. The node gets the list from the exchange and refreshes it each minute. | | size | auto, or a size in pixels. auto lets each seller use its default size. An explicit size goes only to the sellers that declared it. | | seconds | auto, or the length of the video in seconds. auto uses the shortest length that a seller declares for the model. | | cap_usd | The most that one run of the node can cost, in USD. | | start_image | Optional. The first frame of the video, as one image. Only the models that read an image input can take it. Use an image with the shape of size, because a seller can crop a different shape. | | retention_policy | none (the default) excludes providers that can keep your content. allowed permits those providers. |

Select allowed only if you accept that the provider can keep your prompt, start image and video. The node sends this choice as exchange.retention_policy. Old workflows use none until you change it.

A video is priced for each second. Thus the charge is the price of the length that you ask for, and the cap must be at least that price.

Outputs

| Output | Description | | --- | --- | | video | The video file, as a ComfyUI VIDEO. Connect it to Save Video, or to Get Video Components for the frames. | | job_id | The exchange's ID for the job. | | receipt | JSON with the job ID, the model, the seller, the charge and the cap in USD, the length, the size, the expiry of the file and the node version. |

Stop a run

A stop in ComfyUI stops the wait, not the job. The exchange has no route to cancel a video job. The seller continues, and the exchange charges for the video when the job is complete. The error gives the job ID.

To get the video of a job that you paid for, use the Liquid Video Job node:

Add a Liquid Video Job node.
Paste the job ID into job_id.
Connect video to Save Video, and queue the workflow.

This node buys nothing. It needs a key that can read the account's jobs. The seller keeps the file until the time in expires_at in the receipt.

Privacy

The prompt goes to the Liquid Inference router, and then to the seller that wins the auction. The seller can be a different company for each run.
The images come back from the seller. If the seller returns a link and not the image bytes, the node gets the image from that link. The node does not send your key to that link.
A video comes through the exchange. The exchange does not keep the file.
Anyone who can open your ComfyUI server can queue paid runs with your key. Do not start ComfyUI with --listen on a network that you do not trust.

Retries

The paid nodes never send a purchase again by themselves. A second request could buy the images or the video a second time. After the purchase of a video, the node tries a failed status read or download again, because these requests buy nothing.

If the connection stops before the answer to a purchase comes back, the node shows an error that says that the outcome is unknown. The job possibly ran, and your account possibly paid for it. The error gives an Idempotency-Key. Give it to support to find the job.

Troubleshoot

| Error | What to do | | --- | --- | | No Liquid Inference API key | Set the key in Settings, or set LIQUID_API_KEY, then try again. | | no_eligible_offer with retention_policy | For a video, select retention_policy = allowed only if you accept content retention. Otherwise, select a different model. | | HTTP 401 | The exchange does not know the key. Make a new key and save it. | | HTTP 402 insufficient_balance | Add funds to your account. | | HTTP 400 unsupported_parameter | No offer for the model accepts a value that you sent, usually size. Set size to auto, or select a different model. | | HTTP 503 no_eligible_offer with cap_exceeds_buyer_cap | Each offer costs more than cap_usd. Increase cap_usd. | | HTTP 503 no_eligible_offer, other reasons | No seller can take the request now. Try again later, or select a different model. | | HTTP 404 model_not_found | No seller lists the model now. Reload the page to get the current model list, then select a model. | | HTTP 503 model_unavailable | The model has sellers, but none of them can take a job now. Try again in a minute, or reload the page and select a different model. | | HTTP 503 provider_failed | The seller failed before it made an image, and you were not charged. Run the node again. If no other seller has the model, the next run gets model_unavailable. | | HTTP 502 or 504 with a web page | A proxy between ComfyUI and the exchange answered. Look at your receipts before you run the node again. | | The model list shows "no image models" or "no video models" | ComfyUI could not get the model list. Check the network and the base URL, then reload the page. | | A video job failed | The seller did not make the video, and you were not charged. Run the node again, or select a different model. | | The node stopped waiting for a video job | The job continues, and the exchange charges for it. Put the job ID in a Liquid Video Job node to get the video. | | HTTP 410 video_expired | The seller deleted the file. Look at the job's receipt for the charge. | | The outcome is unknown | Do not run the node again at once. Look at your receipts, or give the Idempotency-Key to support. |

Each refusal from the exchange includes the code and the reasons that the offers were excluded. A refusal before the auction costs nothing.

Base URL

The node sends requests to https://router.inference.ai.exchange/v1. To use a different router, set Base URL in Settings or set the LIQUID_BASE_URL environment variable.

Development

The project uses uv.

uv run pytest            # unit tests, with a local stub router
uv run ruff check
uv run ruff format --check
uv run mypy

The smoke test installs a pinned ComfyUI release and CPU torch, starts ComfyUI against the stub router, and runs each example workflow. Then it makes sure that no saved image, no workflow and no server answer holds the key. The first run takes some minutes.

uv run pytest -m smoke

To keep the ComfyUI installation for the next run, set LIQUID_SMOKE_DIR to a folder.

Contribute

This repository is a copy of the source, which is not public. Each change comes from the source, and a tag v<version> publishes that version to the Comfy Registry. Open an issue or a pull request here. A maintainer applies an accepted change in the source, and the next copy includes it. Do not expect a merge of the pull request itself.

License

MIT. See LICENSE.