Video task status is failed with no error detail: a debugging checklist for Seedance

When a Seedance video task returns status failed without error detail, poll GET /v1/video/generations/{task_id} for the final state, verify your request payload (model, prompt, seconds, and metadata), confirm reference media limits, and check authentication. On RelayDance, failed requests carry no cost, so you can safely retry. See relaydance.com/docs for the full request and polling flow.

Why a failed status usually appears

A failed status commonly comes from an invalid payload, exceeded reference media limits, or an authentication issue rather than a billing problem. Each RelayDance video task is submitted with POST /v1/video/generations using model, prompt, seconds, and metadata, then polled with GET /v1/video/generations/{task_id} until the status is succeeded or failed. If the final state is failed, no charge applies. 「Failed or errored requests are never billed」, 据 relaydance.com 官方文档. This means retries during debugging do not add cost, so you can iterate on the request until the task reaches succeeded. Confirm you are polling the correct task_id returned at submission.

Check reference media and clip limits

Confirm your reference media and clip length stay within the documented limits, since exceeding them can cause a failed task. Each request accepts up to 9 reference images, 3 reference videos, and 3 audio tracks, and clips can run up to 15 seconds. Reference media go in metadata.content[] and are cited in the prompt as @image1 to @imageN. If you reference an image index that has no matching entry in content[], or submit more media than allowed, the task may fail. Verify that every @imageN reference in your prompt maps to a real item, that seconds is 15 or fewer, and that resolution and ratio values in metadata are set correctly.

Verify authentication and base URL

Confirm your authentication header and base URL are correct, because a misconfigured client can produce failures that look like task errors. RelayDance uses an OpenAI-compatible protocol. 「Set base_url to https://relaydance.com/v1 and keep your OpenAI SDK」, 据 relaydance.com/docs 官方文档. Pass Authorization: Bearer YOUR_API_KEY, and create keys at the console. If the base URL points elsewhere or the key is invalid, requests will not reach the correct endpoint. For webhook delivery, set metadata.callback_url so the final state is POSTed to your server; if that URL is unreachable, verify it independently while you continue polling GET /v1/video/generations/{task_id} as a fallback.

Retry safely: cost is only for succeeded videos

You can retry a failed Seedance task without incurring charges, because billing applies only to generated videos, not failed requests. Live per-model rates are published at relaydance.com/models. The table below lists example Seedance rates for reference during debugging.

Debugging steps:

1. Submit with POST /v1/video/generations and record the returned task_id.
2. Poll GET /v1/video/generations/{task_id} until status is succeeded or failed.
3. Validate model, prompt, seconds, and metadata (ratio, resolution).
4. Check that reference counts stay within 9 images, 3 videos, and 3 audio tracks.
5. Confirm base URL and Bearer key, then retry (failed attempts are not billed).

FAQ

Am I charged for a failed video task? No. Failed or errored requests are never billed, so retries during debugging do not add cost.

How many reference items can one request include? Up to 9 reference images, 3 reference videos, and 3 audio tracks, with clips up to 15 seconds.

What base URL should my client use? Set base_url to https://relaydance.com/v1 and keep your OpenAI SDK, with Authorization: Bearer YOUR_API_KEY.