Public beta
Wan 3.0
Alibaba announced public beta testing in August 2026.
Wan 3.0 guide · Updated 17 August 2026
Wan 3.0 AI Video Generator
Explore what Wan 3.0 can do, watch real generation examples, copy practical prompts and learn a simple workflow for creating your own videos. When inspiration strikes, open AIWan and start generating with the Wan 3.0 experience in your browser.
Text / image / references Officially announced up to 30s Create directly in your browser
Public beta
Wan 3.0
Native clip window
Up to 30s
Reference inputs
Multimodal
Native clip window
Up to 30s
Official launch material states native generation up to 30 seconds per clip.
Reference inputs
Multimodal
Text, image, video, audio, web pages and document-style inputs are supported in the announced workflow.
These details are based on Alibaba's August 2026 Wan 3.0 public-beta announcement. Availability and product settings can vary by platform, so always check the options shown in the creator you are using.
At a glance
What Is Wan 3.0?
Wan 3.0 is Alibaba's latest publicly tested generation model in the Wan creative family. Alibaba announced the public beta in August 2026 with native video generation up to 30 seconds and a broader multimodal reference workflow designed to combine text, images, video, audio, web pages and document-style inputs.
For creators, the most useful details are the ones you can verify and use today. Wan 3.0 availability, API access, output settings and pricing can change during beta, so this guide focuses on officially announced capabilities and the options currently available in the product. If you want to move from research to creation, AIWan gives you a browser-based way to try the available Wan 3.0 workflow.
Start creating
How to Use Wan 3.0
You can get started in three steps: choose how you want to begin, describe the shot you want, then generate and refine the result.
01
Choose a generation mode
Start with text-to-video for a fresh shot, image-to-video when a first frame should anchor the composition, or a supported reference workflow when identity, style, motion or other source material matters.
02
Write the creative brief
Describe the subject, action, environment, camera, lighting, visual style and sound intent. Treat the prompt like a compact shot brief rather than a list of disconnected adjectives.
03
Generate, review and iterate
Choose only the duration, resolution and aspect-ratio options currently exposed by the enabled model. Review continuity and motion, then refine one variable at a time before scaling the shot.
Capabilities
What Can You Create with Wan 3.0?
Use the model around a clear creative job rather than a vague “make it cinematic” prompt. The strongest workflows connect source material, motion intent and camera direction.
Text to video
Turn a written scene direction into a complete moving shot with explicit subject, action, camera and atmosphere instructions.
Image to video
Use a visual anchor to preserve composition and art direction while introducing motion, camera behavior and pacing.
Multimodal references
The announced Wan 3.0 workflow expands beyond text and images to richer combinations of video, audio, documents and web content.
Reference consistency
Official launch material emphasizes tighter continuity for character details, objects, layouts, style and voice across reference-led creation.
Longer camera ideas
Native clips up to 30 seconds can reduce the need to stitch many very short generations when planning continuous action or camera movement.
Creative iteration
Use prompt, reference and duration changes as controlled experiments, keeping successful frames and creative directions ready for the next shot.
See what it can create
Wan 3.0 Video Examples
Watch 11 Wan 3.0 examples across cinematic characters, animation, automotive motion, products and large-scale scenes. Each clip comes with a ready-to-copy prompt idea you can adapt for your own project.
30s1920×1080
Fantasy Ruins — Cinematic Environment & Character Motion
A wide fantasy scene with layered depth, dramatic ruins, environmental detail and multiple characters moving through the frame.
Prompt idea
Epic cinematic fantasy ruins at blue hour, two adventurers moving through a moss-covered stone sanctuary, volumetric light, deep environmental detail, slow controlled tracking camera, realistic cloth and foliage motion, dramatic scale, filmic contrast.
15s1920×1080
Cinematic Portrait — Natural Expression & Eye Contact
A close-up portrait study focused on facial detail, subtle expression, shallow depth of field and controlled camera presence.
Prompt idea
Intimate cinematic close-up of a fictional young adult in soft daylight, subtle natural smile, realistic skin texture, gentle eye movement, shallow depth of field, restrained dolly-in, neutral interior, premium film look.
15s1920×1072
Beach Product Shot — Bright Lifestyle Motion
A playful outdoor product-style clip with bright daylight, hand interaction, saturated colour and lightweight motion.
Prompt idea
Sunny beach lifestyle commercial, close-up hands opening a bright blue bubble toy, sparkling sea in the background, clean summer palette, crisp natural light, playful macro details, soft handheld movement, cheerful advertising tone.
20s1280×720
Stylized Character — POV Fridge Interaction
A stylized character interaction framed from inside a refrigerator, combining expressive animation with a strong point-of-view composition.
Prompt idea
Whimsical stylized 3D character opens a refrigerator and reaches for a dessert cup, camera placed inside the fridge, expressive face, believable hand motion, colourful food props, playful wide-angle perspective, clean animation lighting.
35s source reel1280×720
Rainy Supercar — High-Speed Camera & Reflections
A night-driving action shot that emphasizes wet-road reflections, vehicle motion, atmosphere and a dynamic chase-camera feel.
Prompt idea
Aggressive futuristic supercar accelerating through a rain-soaked city at night, low rear tracking camera, bright reflections on wet asphalt, tire spray, cinematic motion blur, controlled lane movement, neon ambience, high-energy automotive commercial.
36s source reel1280×720
Animated Adventure — Storybook Motion & Scale
A stylized animation sequence with a childlike protagonist, oversized props and storybook composition suited to narrative concepting.
Prompt idea
Hand-painted animated adventure, small fictional explorer climbing through a surreal world of giant glass bottles and wooden structures, warm golden light, playful scale, expressive body movement, gentle parallax camera, storybook texture.
30s1440×1080
Dramatic Dialogue — Performance-Focused Close-Up
A dramatic portrait shot centered on reaction, dialogue timing and stable facial presentation in a cinematic frame.
Prompt idea
Cinematic dramatic conversation close-up of a fictional adult woman, controlled breathing and emotional reaction, cool window light, realistic facial micro-expression, subtle head motion, shallow focus, steady medium-close framing.
14s1920×1080
Dinner Scene — Two-Character Continuity
A two-person narrative setup testing composition, eyelines, subtle performance and continuity inside one shared scene.
Prompt idea
Two fictional adults seated at a quiet dinner table, restrained emotional tension, warm practical lighting, natural eyelines, subtle hand movement, balanced two-shot composition, slow cinematic push-in, realistic contemporary drama style.
30s1920×1072
Sci-Fi Detail — Object Continuity & Environmental FX
A cinematic object-detail shot using rain, debris and distant fire to create a dense science-fiction atmosphere around a hero prop.
Prompt idea
Macro cinematic shot of a mysterious mechanical device lying on rain-soaked rubble, distant fires and smoke, water droplets, subtle indicator lights, shallow focus, slow lateral camera slide, realistic disaster-film atmosphere.
30s1926×1074
Creature Sequence — Large-Scale Cinematic Staging
A large-scale urban creature scene combining a foreground character, destruction, vehicles and background action in one cinematic composition.
Prompt idea
Cinematic city disaster sequence, fictional adult walking calmly toward camera while a giant shadow creature moves through the damaged street behind, smoke, traffic, practical debris, deep perspective, controlled handheld camera, realistic blockbuster mood.
30s1920×1072
Sunny Lifestyle Shot — Wide-Angle Character Motion
A bright lifestyle frame with an unusual low camera angle, strong daylight and a relaxed character performance in motion.
Prompt idea
Sunny California-style street, fictional young adult relaxing in a red shopping cart, tall palm trees, wide-angle low camera, natural midday light, playful road-movie energy, smooth forward motion, warm nostalgic colour grade.
Prompt library
Best Wan 3.0 Prompts
Prompt formula: Subject + Action + Environment + Camera + Lighting + Style + Audio intent. Add only the elements that materially change the shot.
Cinematic
Cinematic tracking shot
Cinematic tracking shot of a fictional traveler in a red coat walking through a rain-washed future city, camera slowly closes the distance, storefront light reflects across the pavement, natural footsteps, layered fog, realistic fabric movement, restrained teal-and-amber grade.
Use this prompt on AIWanAdvertising
Premium product reveal
Premium studio product film of a fictional matte-black wireless speaker on dark stone, narrow rim light, slow 120-degree orbit, subtle dust in the beam, precise surface reflections, clean background, final centered hero frame, sophisticated commercial pacing.
Use this prompt on AIWanImage to video
Image-to-video portrait
Preserve the source portrait, face, clothing, palette and background layout. Add gentle breathing, a small turn toward window light and a restrained dolly-in. Keep expression natural, hair movement subtle and depth of field consistent with the source image.
Use this prompt on AIWanCharacter
Character continuity shot
Keep the fictional protagonist's face, hairstyle, jacket, body proportions and colour palette consistent across the full shot. The character exits a train, looks left, then walks toward camera while the camera tracks backward at matching speed.
Use this prompt on AIWanCamera motion
Automotive action
Low rear three-quarter tracking shot of a futuristic sports car on a wet mountain road at dusk, controlled acceleration, water spray, accurate wheel rotation, stable body shape, camera remains close to the road, smooth curve into a wide scenic reveal.
Use this prompt on AIWanAnimation
Stylized animation
Warm hand-painted animation of a small fictional explorer crossing an oversized kitchen at sunrise, playful perspective, expressive body motion, soft squash-and-stretch, layered parallax, textured brushwork, gentle camera pan, whimsical family-film tone.
Use this prompt on AIWanSocial media
Social product demo
9:16 vertical product demo, adult hands open a fictional skincare bottle on a bright bathroom counter, clear three-beat structure: package reveal, texture close-up, clean final hero shot; daylight, crisp macro detail, smooth quick camera transitions.
Use this prompt on AIWanNature
Atmospheric nature shot
Aerial cinematic shot over a mist-filled alpine lake at dawn, slow forward drone movement, pine trees emerging through fog, sunlight gradually warms the valley, realistic water surface, subtle birds in the distance, calm documentary pacing.
Use this prompt on AIWanCreative workflows
Wan 3.0 Use Cases
01
Filmmaking & previs
Test camera language, blocking, atmosphere and VFX concepts before committing to production.
02
Advertising
Prototype product films, campaign ideas and visual directions with repeatable prompt structures.
03
Social media
Create short concept clips for vertical, square and landscape content experiments.
04
E-commerce
Turn product stills and creative direction into motion concepts for listings and campaigns.
05
Animation
Explore stylized characters, worlds and transitions without limiting the workflow to photorealism.
06
Education & demos
Transform structured information and visual references into clearer motion-led explanations.
Choose the right model
Wan 3.0 vs Other AI Video Models
Different AI video models are strong at different jobs. Use this overview to narrow your choices, then test the same source material and creative brief in the models that matter most to your workflow.
| Model | Best fit | Reference workflow | Audio | Access |
|---|---|---|---|---|
| Wan 3.0 | Longer multimodal creative workflows | Strong reference-led focus | Multimodal audio input / voice continuity in announced workflow | Public beta / availability can vary |
| Seedance | Multimodal creative production | Version-dependent | Version-dependent | Hosted / API availability varies |
| Veo | Cinematic generation workflows | Product/version-dependent | Supported in selected current workflows | Google product/API availability varies |
| Kling | Motion-heavy creator workflows | Available in selected workflows | Product/version-dependent | Hosted/API options vary |
| LTX | Fast iteration and open/local workflows | Workflow-dependent | Model/version-dependent | Hosted and open-weight paths exist by version |
Model capabilities evolve quickly. For a fair comparison, use the same prompt or reference image and judge motion, consistency, camera control, generation time and cost side by side.
Plans & credits
Wan 3.0 Pricing on AIWan
AIWan uses credits, and you can see the estimated credit cost before you start a generation. Your final cost depends on the model and settings you choose, so check the creator and pricing page for the latest available options.
Basic annual equivalent
$13.9/mo
Pro annual equivalent
$27.9/mo
Max annual equivalent
$90.9/mo
Plan prices and generation costs can change. The live AIWan pricing page always shows the latest plan details, while the creator shows the credit estimate for your selected generation settings.
Ready to make your own video?
Open AIWan, choose an available Wan 3.0 workflow, add your prompt or reference, review the credit estimate and generate when you are ready.
Wan 3.0 API & Developer Access
If you are building with Wan 3.0, start with the current official Wan/Alibaba developer documentation. Beta access and API capabilities can vary by region, account and release stage.
Before integrating, confirm the active model identifier, authentication method, supported inputs, job lifecycle, output handling, error responses and pricing. This helps you build against capabilities that are actually available to your account today.
Developer checklist
- • Verify the active Wan 3.0 model ID and region before deployment.
- • Treat video generation as asynchronous unless the current API explicitly says otherwise.
- • Validate supported input formats and reference limits at request time.
- • Store output durably if provider URLs are temporary.
- • Surface actual cost and failure/refund behavior to users before generation.
Common questions
Wan 3.0 FAQ
Answers to the questions creators most often ask before trying Wan 3.0.
What is Wan 3.0?
Wan 3.0 is Alibaba's latest publicly tested generation model in the Wan creative family. Alibaba announced its public beta in August 2026, highlighting longer native video generation and richer multimodal reference inputs.
What can Wan 3.0 do?
The announced Wan 3.0 workflow is designed for video creation with text and multimodal references, including images, video, audio and document-style inputs. Exact modes available to you can depend on the platform and account access you use.
What is new in Wan 3.0?
The most clearly announced changes are native video clips up to 30 seconds, broader multimodal reference input, intelligent duration, video extension and tighter visual/reference continuity.
How to use Wan 3.0?
Choose a supported generation mode, describe the subject and motion, add any allowed reference material, select only the settings exposed by your current provider, then generate and iterate. AIWan links the creative workflow from this guide to its browser-based studio experience.
How to access Wan 3.0?
Alibaba describes Wan 3.0 as a public beta with testing access through its ecosystem. Availability, regions and account permissions can change. If you want to try it in the browser, AIWan provides access to the Wan workflow currently enabled on the platform.
How do I write better Wan 3.0 prompts?
Write prompts in shot order: subject, action, environment, camera, lighting, style and audio intent. Use concrete motion verbs and camera directions, and avoid changing too many visual variables in one iteration.
How do I create videos with Wan 3.0 from an image?
Use an image-to-video or supported reference mode, keep the source image as the visual anchor, then describe only the motion, camera behavior, pacing and environmental changes you want to introduce.
Is Wan 3.0 free?
Free access depends on the service, beta entitlement and current account policy. AIWan uses credits and shows current plan information separately, so check the live product before generating.
Is Wan 3.0 open source?
Earlier Wan releases include open repositories, but that does not automatically mean the Wan 3.0 beta is available as open weights. If local deployment matters to you, check the current official Wan model repository before planning your setup.
Is Wan 3.0 better than Seedance?
There is no single winner for every workflow. Wan 3.0 is notable for the announced 30-second native window and broad reference inputs, while Seedance capabilities depend on the exact version and access path. Compare the model on your own reference material, motion style, output needs and cost.
Does Wan 3.0 generate audio?
Alibaba's launch material describes text, image, video and audio as simultaneous reference inputs and highlights natural multilingual voice output and voice consistency. The exact audio controls exposed by a third-party interface can differ.
Does Wan 3.0 have an API?
API availability can differ by platform, region and beta entitlement. Check the current Alibaba/Wan API and Model Studio documentation to see which Wan 3.0 capabilities, parameters and regions are available to your account.
Does Wan 3.0 have a watermark?
Watermark behavior depends on the service and plan delivering the output, not only the underlying model name. Check the current export policy of the platform you use before publishing commercial assets.
Does Wan 3.0 support commercial use?
Commercial-use rights depend on the current service terms, account plan, source-material rights and model/provider policy. Review the live terms before using generated output in paid campaigns or client work.
Can Wan 3.0 generate longer videos?
Alibaba's August 2026 announcement states that Wan 3.0 supports native clips up to 30 seconds and also introduces video extension. If you need longer output, check the current product options because available extension limits can vary by release and platform.
Can Wan 3.0 maintain character consistency?
Wan 3.0 is explicitly positioned around stronger visual and reference continuity, including character, object, spatial and voice details. Results still depend on references, prompt clarity, shot complexity and the exact beta implementation.
Try it yourself
Create Your First Wan 3.0 Video
Copy one of the prompts above or bring your own idea, add a source image if your workflow needs one, and create your first Wan 3.0 video on AIWan. You will see the available modes and credit estimate before generating.
Wan 3.0 Information Sources
Want to check the original model information or current product experience? These links take you to official Wan/Alibaba references and the AIWan Wan 3.0 creator.