Manage Hola Flow Queues, Results and Downloads

Give each batch a clear purpose

Group related prompts under a recognizable name, such as “Product close-ups — landscape” or “Opening scene — video tests.” Keep different models, reference sets or aspect-ratio goals in separate groups when that makes the results easier to compare.

Before starting, check the output count, destination folder and file-naming preferences. A folder for each project is easier to review than a single collection of unrelated downloads. Keep the source prompt list with your project notes.

Read the queue before starting more work

Hola Flow’s Queue shows batches and their progress. Inspect a batch’s prompts to understand which items still need attention. Use the interactive preview below to explore sample batches before doing this with your own project.

When progress appears slow, check the active Flow project and any visible error. Waiting for an accepted job is different from submitting a new one. Repeatedly starting the same prompt can create extra generations if the first request was already accepted.

Review the actual result

Open Results and inspect individual images or videos. Check the subject, framing and output type against the original prompt. A successful generation status means the service produced something; it does not establish that the result meets your creative goal.

Separate creative revisions from technical retries. Change the prompt when the composition is wrong. Investigate the connection or download status when the expected file is absent. This keeps a small problem from turning into a second full batch.

Confirm that files reached your computer

Check your save location and Chrome’s download history. A generated preview can exist before a local download finishes. If Flow already contains the completed media, try recovering or downloading that item rather than regenerating it solely to obtain the file.

For videos, open the saved file and confirm it plays. For images, check the dimensions and appearance you need for the destination. Keep the best outputs and a record of their prompts so later revisions are easier.

Pause and diagnose repeated failures

If multiple items fail in the same way, stop adding work and read the error. Confirm that the Google account is still signed in and the intended project is open. Google account limits, model availability and rejected inputs require different responses; increasing retries is not a general solution.

When asking for support, include the extension version, generation mode, error text and whether the item already exists in Flow. Remove private account information. For a cleaner next batch, return to the prompt-planning guide.