Unboxing a product
A close-up of hands opening a small cardboard box. The camera is directly above. The box lid is lifted to reveal a shiny object. Bright, even lighting.
Is the unboxing motion clear? Is the object visible inside the box?
Wan 3.5 AI Video Generator for Social Clips
Draft a short video opener that holds attention on a phone screen. Describe the subject, pick the mobile frame, and preview the clip before you add text or arrange the final post.
Starting prompt briefs for social video ideas
These are example briefs to begin a social video opening. They describe a hook and ask you to check clarity, not show a generated result.
A quick recipe step
A top-down view of a hand sprinkling herbs into a bowl of pasta. The camera is static. The herbs fall onto the pasta. Bright kitchen lighting.
Are the herbs visible as they fall? Is the pasta clearly visible?
A before-and-after reveal
Start: a messy desk with papers and a coffee cup. A hand sweeps across the frame. End: the desk is clean and organized. The camera is static.
Does the transition happen quickly? Is the final desk clearly organized?
A fashion detail close-up
A close-up of a hand touching a textured fabric on a jacket. The camera slowly zooms in on the fabric. The background is blurred. Soft studio light.
Is the fabric texture visible? Does the zoom enhance the detail?
What kinds of social openings can you explore with Wan 3.5 Video?
These illustrations show composition ideas you can describe in a prompt. They are concept references, not outputs generated on this site.
Product reveal with intentional negative space
Plan a close shot of a product with space around it. Describe the object and one subtle movement, then check the label and edges before captioning.
Fashion poster styled for social framing
Use the VELVET DROP poster to set up a balance between person, background, and headline. Request a separate motion from that composition, then examine clothing and lettering.
Calm interior with a quiet focal moment
Base a restrained shot on the vase-and-books still life. Describe the light and one camera move, then watch how objects and frame edges behave.
Material mood board for post direction
Use the four-panel texture board to settle on a color and surface feel before requesting a scene. It's a planning aid; build your post from separately reviewed clips.
How do you turn an idea into a Wan 3.5 Video clip?
Each request answers one visual question. Describe the scene, submit, and look at what comes back — then adjust the next request based on what you see.
- 01
Define a single opening action
Decide what the viewer should notice first: a product, a face, a change. Put that in the prompt and keep the rest of the scene simple.
- 02
Frame for a phone screen
Pick an aspect ratio from the selected model's options. Describe where the subject sits in the frame, and leave space for captions you'll add later.
- 03
Give the subject one clear action
Write a simple movement and a camera direction. Watch the whole clip, including the last frame, to see if the motion fits your opening idea.
- 04
Finish the post in your editor
Download the clip you want, then add captions, sound, pacing, and branding in your own tool. The workspace doesn't publish posts or build a timeline.
What Wan 3.5 Helps You Build
This section focuses on the useful visual assets you can create for short social videos, from opening hooks to main action sequences.
- Scroll-stopping opening frames
- Generate a compelling first frame that defines your video's hook and sets the visual tone for the main content.
- Core action moments
- Create the key action sequence or central visual moment that carries your short video's narrative or message.
- Mobile-optimized compositions
- Produce visuals with composition and framing suited for vertical viewing and small screen engagement.
- Assets for Exploring a Visual Style
- Use references to explore a cohesive visual style across different shots or scenes within your video concept.
Wan 3.5 Video: Your Text and Reference Shape the Generated Concept
Wan 3.5 translates your written description into a short video concept, and can also incorporate visual cues from a supported reference image. The text sets the scene, action, and mood, while a reference provides style or composition direction. The final output is a generated draft for social opening moments, not a finished edit.
Text Describes the Scene
Your written prompt is the primary instruction set. It tells the generator what to show, what is happening, and the desired visual perspective.
- Define the subject, setting, and key objects in the scene.
- Describe the main action or movement sequence you want.
- Specify composition requests, such as "wide shot" or "dynamic angle."
Reference Guides the Style
An optional visual reference image helps align the generated output with a specific aesthetic or mood you have in mind.
- Influence the color grading and overall lighting feel.
- Suggest the texture and style of the environment or characters.
- Provide a visual anchor for the desired atmosphere.
Composition and Motion
You can request specific camera movements or subject actions within your prompt to guide the generated video's flow.
- Request actions like "walking towards camera" or "product spinning."
- Suggest camera moves such as "slow zoom in" or "tracking shot."
- The generator interprets these instructions to create a coherent motion sequence.
Interpreted Output
The final video is a creative interpretation of your inputs. It captures the essence of your description and reference, but exact details like text, precise typography, or guaranteed sequence consistency are not preserved.
- The output is a starting point or concept for your further editing.
- Descriptive text sets the core scene, subject, and intended action. It defines the creative direction for the opening moment you want to build.
- A supported visual reference guides style, composition, or specific visual elements. It helps anchor the generated result to a particular look or mood.
How do credits affect your Wan 3.5 Video workflow?
Test competing hooks—for example, a quick unboxing versus a before-and-after reveal—in separate requests before investing in decorative polish.
Use the balance visible in your account
Sign in to see the credits available to that account. An existing balance is required for generation while new purchases remain paused.
Review the estimate after choosing settings
Model and output choices can change the displayed credit cost. Read the estimate for the current request instead of assuming another model or earlier attempt has the same cost.
Keep each attempt focused
Give one request a clear review goal. Inspect the result before deciding whether another attempt is useful, and keep the details of the brief you want to revisit.
Check the recorded task and balance
If a request fails or its status is unclear, review the task status and account balance before resubmitting. Contact support with the task reference if the outcome needs investigation.
What should you know about Wan 3.5 Video?
Answers about the site brand, how prompts and references work, what to check in a result, and current credit availability.
Site identity and model selection
Understand the difference between the website brand and the actual model chosen for a request.
References and result review
Check which inputs are supported and inspect the returned details before using an output.
Planning and external editing
Use separate requests for your planned assets and finish assembly or detailed editing externally.
Credits, checkout, and examples
Understand the paused purchase flow and the role of the concept illustrations.
Create a social video opening
Write a brief describing subject, action, and framing for vertical or square format. The model generates a short clip for evaluating the hook and mobile composition.
Shape your next video concept with Wan 3.5 Video
Create a short video concept that opens clearly and reads well on a phone. Describe the subject, choose the action, pick a crop, then review the result before you add captions or assemble your post. Check your account balance and the selected model before submitting.