Vidu Q1 (Video)

Vidu Q1 is Shengshu's first-generation quality-focused video model, and the Runware node for it has a personality you'll notice immediately: no required inputs, a prompt box that's fully optional, and a style dropdown sitting front and center. That dropdown is the giveaway - Q1 is the model you pick when you care more about how the video looks than about nailing a specific action. It's the mood piece of the Vidu lineup, and this node hands it to you without a single local dependency.

The workflow the node clearly wants: pick a style from the dropdown, maybe drop in a starting frame via frameImages, optionally write a prompt to steer the action, and out comes a 5-second clip. If you skip the prompt entirely it still works - the style plus any frame you provide does the heavy lifting. For someone who wants video results without assembling the local ComfyUI video stack, that's about as frictionless as it gets.

How it works

Same cloud machinery as every node in the pack: the node (vidu:1@1, videoInference) builds a request with your widgets, sends it over REST through the runware-sdk, and returns a native VIDEO output downloaded from the result URL. frameImages is the one media socket - actual starting frames, uploaded automatically by the SDK. Duration defaults to 5 seconds, and like all the Vidu nodes, frames = duration × the model's fps, with no separate frame-count control.

The inputs that matter

• providerSettings.vidu.style - the star of the show. A visual-style dropdown (default general); this is what separates Q1 from the plain Vidu 2.0 node. Want a particular look? This is where you ask for it.
• frameImages - starting frame(s); strongly recommended if you want motion that respects a specific subject or scene.
• positivePrompt - optional; describe movement and composition.
• duration - default 5 seconds from the preset list.
• providerSettings.vidu.bgm - toggle for background music.
• providerSettings.vidu.movementAmplitude - auto by default; override for more or less motion.

The rest is the pack-standard tail: width/height, seed, numberResults, outputFormat (MP4), outputQuality, safety, and ttl.

Install and API key

The install is the pack's one-size-fits-all:

cd ComfyUI/custom_nodes
git clone https://github.com/Runware/ComfyUI-Runware
pip install -r ComfyUI-Runware/requirements.txt

Restart ComfyUI (or install Runware from ComfyUI Manager). Grab an API key at runware.ai/api-keys and either paste it into ComfyUI Settings → Runware API key, export RUNWARE_API_KEY, or run runware auth login.

Gotchas

Because the prompt is optional, beginners sometimes assume it's broken when results drift from what they vaguely wanted - it's not; you just haven't told it anything. Write the prompt and set a style and you'll actually see what Q1 can do. Two other things: the style dropdown's exact entries are whatever Runware's catalog defines, so treat it as a list to browse rather than a spec sheet. And remember you're paying per clip - five seconds at a time adds up, so settle on a style before you start generating variations. If your output feels over-stylized, drop back to the generic Vidu 2.0 node rather than fighting the style knob.