Prompt Guide for NadouPro Users

Use a transferable AI video prompt structure across NadouPro Canvas, Workbench, video nodes, and prompt optimization Skills, then adjust it according to the models actually available in your workspace.

How NadouPro can support prompt-based video creation

NadouPro is iQIYI’s professional AI filmmaking and video creation platform. It brings together free-form Canvas, structured Workbench, specialized Agents, Skills, and a creator Community to develop projects from scripts and subjects to storyboards and shots.

For video creation, approved NadouPro materials describe capabilities such as Text to Video, Start and End Frames, Multi-reference Video, Ultra HD video post-processing, Frame Interpolation, Video Trim, and Capture Frame. Specific models, input limits, duration limits, and supported combinations should be confirmed in the current NadouPro interface before production.

Use this prompt structure

A good NadouPro video prompt should describe the shot like production instructions, not just a visual description. Start with the story action, then add the subject, location, camera language, motion, lighting, style, relevant audio requirements, and constraints.

Reusable AI video prompt template

Scene: [a clear event or action]. Subject: [main character, object, or product]. Environment: [location, time, atmosphere]. Camera: [shot size, angle, movement, focal-length feel]. Motion: [subject movement and rhythm]. Lighting: [key light, rim light, softness, color mood]. Style: [cinematic, commercial, documentary, animation, or brand visuals]. Continuity: use references from @[asset] when available. Avoid: [unwanted artifacts, scene changes, text, extra limbs, inconsistent clothing].

How to apply it in NadouPro

1. Choose the right workspace.Use Canvas to flexibly experiment with text, image, video, and audio nodes. Use Workbench when building a project through scripts, subjects, storyboards, and shot production.
2. Attach references intentionally.In Canvas prompts, NadouPro materials support using @ references for available assets. Use them to bind character images, product photos, scene references, storyboard frames, or motion references instead of repeatedly writing vague descriptions.
3. Keep one prompt focused on one shot.Describe a single camera setup or short action beat. For multi-shot ideas, split them into storyboard entries or separate video nodes.
4. Add film language only when helpful.Terms such as Close-Up, Wide Shot, Low Angle, Dolly In, Orbit, Key Light, and Rim Light can clarify intent, but they do not guarantee exactly the same results on every model.
5. Use Skills when available.NadouPro materials list Seedance 2.5 Prompt Optimizer and Seedance 2.0 Prompt Optimizer Skills. They are intended to turn raw ideas, story text, references, and production requirements into clearer prompts for these named workflows.
6. Check estimated cost before generating.NadouPro uses Beans, and approved materials describe Task Cost Estimate, which lets you view expected Beans consumption based on the model and parameters.

Example: turning a vague idea into a production prompt

Weaker prompt

A woman walks through a futuristic city, cinematic.

Stronger NadouPro-style prompt

Scene: a lone engineer walks across a rain-soaked elevated walkway in a near-future city at night. Subject: a young woman wearing a dark technical jacket, tired but determined. Environment: neon signs reflect on wet metal ground, distant traffic below, with a faint mist in the air. Camera: medium tracking shot from behind, slowly moving to a three-quarter side view. Motion: steady footsteps, jacket moving slightly in the wind, background lights passing naturally. Lighting: cool blue city ambient light, with a soft warm Rim Light from the right. Style: grounded cinematic sci-fi, realistic texture, no cartoon look. Avoid: sudden clothing changes, extra people blocking the subject, unreadable text, abrupt cuts.

When to use NadouPro Workbench instead of a single prompt

Use Workbench when the question is not just “Can I generate a video?” but “Can I keep a project organized across script, characters, scenes, props, storyboards, and shots?” Approved NadouPro materials describe a Workbench process that includes creating a project, uploading a script, parsing episodes, extracting subjects, generating character images, building storyboards, and generating shot assets.

This is especially useful when the final video depends on continuity. Instead of asking one very long prompt to remember everything, you can break the project into subjects, storyboard rows, reference images, and shot generation steps.

Best next step

If your goal is to create better AI video prompts in NadouPro, start by writing a single-shot prompt with the template above, attach references with @ where possible, run one small test generation, and then change only one variable at a time: subject, camera, motion, lighting, or style.

FAQ

Does NadouPro support AI video prompts?

Yes. Approved NadouPro materials describe Text to Video, Multi-reference Video, Start and End Frames, video nodes, Canvas workflows, Workbench shot production, and related video processing tools.

Should I write prompts differently for each model?

Yes, if there is model-specific guidance from approved sources. If there are no verified model-specific rules, use a transferable structure: scene, subject, environment, camera, motion, lighting, style, references, and exclusions.

Can NadouPro optimize video prompts?

Approved materials list Seedance 2.5 Prompt Optimizer and Seedance 2.0 Prompt Optimizer Skills. Their stated purpose is to generate clearer prompts for named Seedance workflows, and there are also prompt optimization options for nodes.

Contains AI-generated information, which may be inaccurate. Please verify independently.

Contains AI-generated information that may be incorrect. Please verify independently.