RH Seedance2.0素材/查询
Check your Seedance asset actually landed before you spend a generation on it
- asset_id
- status
- preview_url
- response
The whole point of RunningHub's asset system is that you upload a character reference once and reuse it across Seedance generations. But uploads can silently fail - a rejected file, a moderation flag, a network hiccup - and if you feed a dead asset ID into a video generation, you've spent money on a call that was never going to work. DapaoRHSeedanceAssetQueryNode ("RH Seedance2.0素材/查询") is the cheap insurance policy: it asks the platform what state your asset is in before you commit a paid generation to it.
What it does
Three inputs that matter: 🌐 API渠道 (国内版/国外版 - match your key), 🔑 API密钥, and asset_id, the string your 素材/创建 node returned. Give it those and it hits RunningHub's asset endpoint and reports back:
- asset_id - echoed back so you can chain it onward.
- status - this is the one to watch. The pack treats states like ACTIVE/SUCCESS/READY as ready-to-use; FAILED/ERROR/CANCEL means it didn't land.
- preview_url - a link to look at the uploaded material, which is genuinely useful for confirming you uploaded the right face and not a blurry frame.
- response - the raw platform payload when you need the details.
There's also 🚫 跳过错误 and ⌛ 请求超时 (default 60s - queries are fast, no need to inflate it).
The workflow it slots into
Used properly, it's the middle of a three-node chain:
RH Seedance2.0素材/创建 → RH Seedance2.0素材/查询 → RH Seedance2.0素材ID/合并 → RH seedance2.0-Mini
↑
check status is ready here
Realistically, most people don't query on every single run - you check once when you first register a character, confirm it's active, and then trust it. Where this node earns its keep is batch pipelines and debugging: when a generation comes back without the expected character consistency, querying the asset is the fastest way to find out whether the problem is the generation or the material. It's also a nice sanity gate in an automated loop - feed the status back and only proceed on a ready state.
Install
Nothing unusual. This whole pack installs one way:
cd ComfyUI/custom_nodes/
git clone https://github.com/paolaoshi/ComfyUI-dapaoAPI.git
cd ComfyUI-dapaoAPI
pip install -r requirements.txt
Or ComfyUI Manager → search dapaoAPI → Install → restart. No model files, just your RunningHub key.
One honest limitation: the query only tells you what the platform says about the asset, not what the asset actually looks like - that's what preview_url is for, and it's worth a glance. And the asset system is RunningHub-specific, so this node (like its create/merge siblings) is only useful if you're on the RH Seedance path in the first place. If you're feeding plain URLs into the mini node instead of asset IDs, you don't need any of the three asset nodes at all. But for character-consistency work, that two-second status check saves real money and real confusion.
Inputs (5)
| Name | Type | Default | Description |
|---|---|---|---|
| 🌐 API渠道 | COMBO | 国内版 | 国内版与国外版使用不同的 API 地址和 API 密钥,请选择与密钥一致的渠道。 |
| 🔑 API密钥 | STRING | RunningHub API Key,仅用于本次请求,不会写入文件。国内版和国外版密钥不通用。 | |
| asset_id | STRING | — | |
| skip_erroropt | BOOLEAN | false | — |
| ⌛ 请求超时opt | INT | 6010–300 | — |
Outputs (4)
| Name | Type | Description |
|---|---|---|
| asset_id | STRING | — |
| status | STRING | — |
| preview_url | STRING | — |
| response | STRING | — |