%22/%3E%3Cpath d=%22M4.5 7h3l1.5 9 2-7h2l2 7 1.5-9h3l-2.8 10h-3.1L12 12l-1.6 5H7.3z%22 fill=%22white%22/%3E%3C/svg%3E)
Try Wan 3.0 free with eligible Topview trial credits. Turn text prompts, images, and supported multimodal references into AI-generated videos through a browser-based creative workflow. Developed by Alibaba, Wan 3.0 supports videos up to 30 seconds, resolution up to 1080P, and synchronized dialogue, music, and sound effects. Choose the available generation mode, add your creative direction, and explore what you can make.
Free trial credits for eligible users · Wan 3.0 720P unlimited generation on select paid plans · Credits, availability, and limits are shown before generation.
@Image 1Explore six creative concepts showing how Wan 3.0 prompts can describe narrative structure, camera movement, character performance, and visual styles. Each example includes a creative brief you can adapt for your own video. The result of a new generation may differ from the showcased concept.
A Robot Finds Spring
A battle-scarred robot walks through ruins reclaimed by nature and discovers a single flower beneath the cherry blossoms—a quiet story told through environment, scale, and emotion.
Create a sci-fi short film about a battle-scarred robot discovering the beauty of spring as flowers and cherry blossoms overtake the ruins of war.
The Moment She Learns the Truth
A betrayed woman silently pieces together what happened before facing the person responsible. Tight close-ups, restrained acting, and controlled lighting turn the reveal into a complete dramatic beat.
Create a live-action revenge drama that begins with a woman discovering the truth and builds toward a cold, satisfying confrontation.
Impact in Every Move
Dynamic tracking, whip pans, suspended impact moments, and a clear final pose give this fight scene the rhythm and physical force of a choreographed action sequence.
Create a high-energy fight scene with dynamic tracking shots, whip pans, slow-motion impacts, and a clean heroic finish.
The Final Sword Strike
A lone swordsman gathers power, releases one decisive attack, and defeats the demon king. The sequence builds tension before delivering a magical impact and victorious ending.
Create a Japanese anime-style scene where a lone swordsman unleashes a final ultimate attack against the demon king.
A Gentle Bedtime Adventure
A soft wool-crafted rabbit explores a bright miniature garden. Handmade textures, warm color, and gentle movement make the clip feel like a children’s storybook brought to life.
Create a warm felted-wool children’s storybook video with a gentle handmade look and a curious rabbit hero.
A World Worth Entering
Hero portraits, armored enemies, world-building shots, and combat moments combine into a cinematic concept trailer that introduces the game before its final reveal.
Create a fantasy action-game CG trailer with the world, main hero, powerful enemy, epic combat, and a dramatic final reveal.
Wan 3.0 is a multimodal AI video model developed by Alibaba's Wan team. Its official capabilities extend beyond basic text-to-video generation to include image-guided video, multimodal references, audiovisual generation, and editing workflows.
Create a story with an opening, development, and ending.
Wan 3.0 officially supports video generation from 2 to 30 seconds. Longer clips can give creators more room for scene development, character actions, and narrative transitions. When planning a longer video, divide the sequence into clear shots or story beats instead of placing every action in one long sentence.
Let the story guide the timing.
Wan 3.0 supports a smart-duration option at the model level. Rather than always requesting the maximum duration, creators can plan a video around the amount of action required. For precise advertising formats, use an explicit duration when the selected interface supports it.
Combine creative instructions with relevant source material.
Wan 3.0 can use images, video, audio, documents, and public webpages as references in supported workflows. A product image can define appearance. A video reference can guide motion. Audio can guide rhythm or performance. A document or webpage can provide narrative context. Not every reference combination is valid, and not every mode exposes the same upload options.
Describe the visual details that matter.
Use prompts to guide lighting, materials, facial expressions, environments, camera angles, and artistic styles. Wan 3.0 can support both realistic and stylized creative directions, including cinematic action, animation, product demonstrations, and fictional worlds. Generated text, logos, anatomy, and physical interactions should still be reviewed before publishing.
Coordinate visual identity, motion, and sound.
Reference-based workflows can help guide the appearance of characters, products, and environments. Wan 3.0 also supports native dialogue, background music, and sound effects. For more consistent results, identify which elements must remain unchanged and which elements are allowed to move or transform.
A clear creative brief helps Wan 3.0 interpret a complex scene. These five capabilities are useful starting points when planning a new generation.
The preview represents a 30-second option for ideas that need an opening, development, transition, and conclusion. Its smart-duration concept shows how explainers, commercials, and narrative scenes could receive more room when the live workflow supports it.
The preview groups image, text, video, and audio direction, with documents or webpages represented as structured context. Use the live Topview generator to confirm which sources the selected model and mode currently accept.
Use the brief to describe skin, materials, light, motion, voices, environments, software UI, typography, and animation. These examples demonstrate the intended direction; generated realism and text accuracy still require frame-by-frame review.
Identify the characters, products, props, voices, locations, framing logic, and spatial relationships that should follow the reference. Treat consistency as a direction to review, not a guaranteed pixel-perfect result.
Use action, sound, acting, camera energy, pacing, and plot in the same creative direction. The example briefs show how to ask for a reveal, tension, performance, or emotional finish when the selected workflow supports the required inputs.
Wan 3.0 in Topview offers workflows that can connect visual generation to information and media supplied by the creator. The following examples show how different sources can help shape an AI video.
Plan a paced visual explanation from a product brief, pitch deck, lesson plan, or report. When document input is available, use it for facts and hierarchy, then add supported image, UI, character, or audio references.
Plan a launch film or social ad around webpage content. When webpage context is available, identify the product language, visual system, interface structure, and key messages that the video should adapt into motion.
Build a connected brief around characters, locations, props, voice, music, and a storyboard. Reuse the supported reference set for variations, then review every output for identity and world consistency.
This Wan 3.0 workflow preview is aimed at teams exploring more narrative space, reference-led direction, and a closer connection between visuals and sound.
Block complete beats, explore camera language, test performances, and communicate a scene before committing to a full shoot.
Keep products, typography, interface details, and campaign style recognizable across launch films, ads, and format variations.
Combine a client's deck, site, brand references, voice, and storyboard into reviewable video directions without rebuilding the brief by hand.
Prototype environments, character interactions, UI overlays, and story moments while maintaining the spatial logic of the world.
Plan visual explanations from documents, software workflows, and structured information, then verify supported inputs and review digital details for accuracy.
Build around a track, voice, rhythm, and performance reference so the edit feels driven by sound rather than added after the fact.
You can start exploring Wan 3.0 on Topview with eligible trial credits. A browser-based workflow lets you provide a prompt, add supported references, and choose available generation settings.
Open the Wan 3.0 generator and select the mode that matches your input material. The mode selector determines which inputs and controls are available.
Describe the subject, movement, setting, camera, visual style, timing, and sound. If you use an image, identify what should remain consistent. For multiple references, name each source and explain its role. Review the editable direction and reference image, check which model and settings are currently available, then start the generation workflow.
When the result is ready, review visual accuracy, motion, and sound. Refine the prompt if necessary, then export the strongest result for ads, social posts, product videos, or creative projects.
Wan 3.0 image-to-video generation turns a still image into a moving video based on a written prompt. The image supplies visual information, while the prompt describes what should happen next. The official model supports both first-frame animation and first-and-last-frame video generation. Check which options are available in the current Topview interface.
A first-frame workflow uses your uploaded image as the opening visual and generates motion from that starting point. This is useful for product photography, portrait animation, illustrated scenes, and environment shots.
When supported, this workflow uses separate opening and ending images to guide the visual transition between them. It can help plan a transformation, camera transition, or movement from one composition to another. First-and-last-frame inputs are a separate task type from general multimodal reference inputs, so do not assume every reference type can be combined in the same generation.
Wan 3.0 is available through different Topview access options. Free trial access, credit-based generation, and Unlimited benefits have different conditions.
Yes. Eligible Topview users can use trial credits to explore supported Wan 3.0 generation workflows. Free trial access is limited. The available allowance, supported settings, and generation costs depend on your account and the current product configuration.
Yes. Topview's published Ultra and Team plans include Wan 3.0 720P 365 Days Unlimited, subject to the current plan terms. Unlimited means eligible credit-free generation through the Topview website during the applicable benefit period. It does not mean that every resolution, API call, or generation method is included. Unlimited tasks may have longer queue times, depending on demand and account activity.
| Access Option | What It Includes | Important Limit |
|---|---|---|
| Eligible free trial | Limited trial-credit generation | Credit and account eligibility apply |
| Credit-based generation | Pay with applicable Topview credits | Cost depends on current settings and plan |
| Ultra Unlimited benefit | Wan 3.0 720P for the stated 365-day period | Subject to plan and queue conditions |
| Team Unlimited benefit | Wan 3.0 720P for the stated 365-day period | Subject to plan and seat conditions |
Unlimited benefits are for eligible in-platform use and do not extend to API, CLI, or other automated generation methods. Always review the latest plan details, generation limits, and billing terms before subscribing.
Topview API
Integrate Wan 3.0 into your application with Topview API. Build text-to-video, image-to-video, and multimodal reference workflows with transparent developer documentation and competitive platform rates.
Loading current Wan 3.0 API rates…
Explore Wan 3.0 APIView API DocumentationReference rate for the configuration shown. Pricing varies by settings and usage. API pricing is separate from Topview's web generator credits.
A useful Wan 3.0 prompt describes more than a visual style. It defines what appears, what happens, how the camera moves, which references matter, and how the video should sound. For longer sequences, a shot-by-shot structure can help communicate timing and transitions.
Subject + Scene + Action + Camera + Timing + Visual Style + Audio + Reference Instructions
For a multimodal project, add explicit references such as Image 1, Video 1, or Audio 1 and explain the purpose of each source. Use only the reference types available in your selected generation mode.
Best for: E-commerce, brand advertising, technology products.
Prompt
Create a 30-second premium commercial for the wireless headphones shown in Image 1. Preserve the product's overall shape, color, button positions, and material finish. Shot 1 (00:00–00:06): Extreme close-up of the headphones against a dark studio background. Soft light reveals the surface texture as the camera moves across the product. Shot 2 (00:06–00:12): Transition to a wider rotating product shot. Use clean reflections and a minimal background. Shot 3 (00:12–00:18): Show a person putting on the headphones while walking through a contemporary city. Keep the product design consistent. Shot 4 (00:18–00:24): Show a calm lifestyle moment on public transport. The person closes their eyes and relaxes while the background becomes visually quieter. Shot 5 (00:24–00:30): Return to a simple product hero shot with restrained studio lighting and an elegant final composition. Audio: Modern electronic music, gentle environmental sound, and subtle transition effects. Avoid adding unprovided logos or invented product claims.
Best for: Short films, storytelling, previsualization.
Prompt
Create a 30-second cinematic scene set inside a nearly empty overnight train during heavy rain. The main character is a woman wearing a dark green coat and holding a small silver case. Keep her face, coat, and the case consistent throughout the sequence. Shot 1 (00:00–00:06): Wide shot of the train moving through a dark landscape. Rain slides across the windows. Shot 2 (00:06–00:12): Inside the compartment, the woman examines a small note and looks toward the locked door. Shot 3 (00:12–00:18): Close-up of her face as she hears footsteps outside. Her expression changes from curiosity to concern. Shot 4 (00:18–00:24): The camera follows her into the corridor. She reaches the end of the carriage and stops. Shot 5 (00:24–00:30): The lights flicker, the silver case opens slightly, and a blue glow illuminates her face. End before revealing what is inside. Audio: Train vibrations, rainfall, footsteps, faint metallic sounds, and a restrained suspenseful score. Use natural camera movement and cinematic low-key lighting.
Best for: Product photography, social media content, visual storytelling.
Prompt
Animate Image 1, showing a vintage red bicycle beside a small flower shop on a rainy street. Preserve the bicycle frame, basket, wheel placement, and overall composition. Begin with a stationary medium shot. Raindrops create small ripples in nearby puddles while flowers move gently in the breeze. Slowly push the camera toward the bicycle as the shop's warm interior light reflects on the wet pavement. Add natural rainfall, distant footsteps, and quiet street ambience. Keep the motion subtle and the original image recognizable throughout.
Best for: SaaS, product demos, internal training.
Prompt
Use the supplied product webpage as the factual reference for a short software explainer. Create a clean, modern visual presentation with clear transitions and minimal motion graphics. Start with the user's main problem, introduce the software interface, then demonstrate three key steps using the source material. Preserve the original feature names and avoid inventing buttons, numbers, or interface options. Use a light neutral background, readable compositions, restrained animations, and a calm English voiceover. End with a summary of the product's main benefit. Verify all interface labels and claims against the supplied webpage.
Alibaba Cloud officially documents Wan 3.0 as a multimodal video generation model supporting text-to-video, image-to-video, multimodal reference input, audiovisual output, and editing workflows. The following specifications describe the model's official capabilities. The actual features available in Topview depend on the selected mode, plan, and current service configuration.
Alibaba's model documentation allows up to 20 multimodal references in a supported request, with specific limits on individual input categories. The total asset limit and valid combinations still apply. First-frame and first-and-last-frame modes cannot be mixed with general reference inputs in the same request.
2–30 seconds
Official support includes fixed lengths and a smart-duration option. Confirm the options shown in the live Topview generator before planning an exact deliverable.
Up to 1080P
Confirm the output tiers available for the selected model and plan before you render.
Text · Image · Video · Audio
Use the live generator to confirm which of these input types the selected model and mode accept.
Documents · Webpages
Document or webpage reference: one of these input types in a supported request, not both together.
Up to 10
Counts toward the overall multimodal reference limit of up to 20 assets in a supported request.
Up to 5 clips
Combined reference-video duration up to 15 seconds.
Up to 5 clips
Combined reference-audio duration up to 15 seconds.
Text · Image · Omni · Edit
Choose from the modes currently available in Topview for the source material you have.
Dialogue · Music · SFX
Wan 3.0 supports synchronized dialogue, background music, and sound effects at the model level.
Shot-level direction
Describe camera, action, performance, pacing, sound, and continuity constraints in one creative brief.
Wan 3.0 gives creators a way to explore longer video sequences, image-guided animation, multimodal references, and audiovisual storytelling.
Free trial credits for eligible users. Generation availability and credit use are shown in the product workflow.
Related Resources
Compare models, review pricing, or open the Wan 3.0 API documentation before you choose a workflow.
