Amber & Stone
Scene prompt: An amber perfume bottle on cream travertine. Slowly push in as warm sunlight creates delicate glass reflections and moving shadows.
Try this promptText to Video AI: Turn Words into Moving Scenes
Describe a scene, camera movement, and sound. Wan3 turns your prompt into a video with native audio, from product close-ups to sweeping landscapes. Start with words; no image upload needed.
Text to Video AI Generator: Explore Scene Prompts
Explore Wan3's existing showcase videos and their scene prompts. Use a prompt as a starting point, then adjust the subject, lighting, and movement in the creator. Results vary with the model and settings.
Alpine First Light
Scene prompt: Glide above a turquoise alpine lake at sunrise. Snow-capped peaks reflect in the water as mist drifts along the evergreen shoreline.
Try this promptThe Espresso Ritual
Scene prompt: Fresh espresso flows into an ivory ceramic cup. Move closer to the golden crema and rising steam in warm side light.
Try this promptHow to Use the AI Text to Video Generator
Write a prompt above and choose Create from my prompt. The creator opens in Text to video mode with your words filled in.
Choose a model, clip length, aspect ratio, and resolution. Describe the sounds you want alongside the visual direction.
Sign in and add credits if needed. Review the displayed credit cost before generating; your prompt carries through the creator's sign-in flow.
Generate, preview the video with its native audio, and download the MP4.
Direct a Realistic AI Video Generator with Specific Details
Describe one clear shot: the subject, what moves, the light, and one camera move. For a natural look, try concrete details such as warm window light, soft reflections, or a slow push-in. Prompt direction guides the result; it does not guarantee a particular scene or level of realism.
Native Audio Generated with Your Video
Sound is generated with the video, including scene ambience and sound effects. Mention sources such as waves, footsteps, or an espresso machine in your prompt. Native audio is always on for W3.0 models and enabled by default for Seedance 2.5 in the creator, with no separate audio charge.
Choose Your Model, Resolution, and Format
- W3.0 and W3.0 Prime: 480P, 720P, or 1080P; clips from 2 to 30 seconds. Prime uses more credits for faster generation.
- W3.0 Pro and Prime Pro: 1080P, 2K, or 4K; clips from 2 to 30 seconds. These models require an active membership.
- Seedance 2.5: 480P, 720P, or 1080P; clips from 4 to 30 seconds.
480P and 720P on non-Pro models work with purchased credits without a membership. All 1080P, 2K, and 4K output requires an active Basic, Pro, or Max membership. Text to video supports 16:9, 9:16, 1:1, 4:3, and 3:4 frames.
Explore More Wan3 Tools
Text to Video AI Pricing
Generation requires credits. Standard W3.0 uses 0.5 credits per second at 480P, 1 at 720P, and 2 at 1080P. Prime, Pro, and Seedance 2.5 costs depend on the selected model and resolution; the creator shows your total before generation. Credit packs do not unlock member-only models or resolutions. Compare plans and credit packs.
Text to Video AI FAQ
What is text to video AI?
Text to video AI generates a video from written scene direction. In Wan3, you can describe a subject, movement, camera angle, lighting, and sound without uploading an image.
How do I use Wan3's AI text to video generator?
Enter a prompt on this page and choose Create from my prompt. Wan3 opens the creator in Text to video mode with the prompt prefilled. Choose your settings, sign in, and make sure you have enough credits before generating. The creator saves your prompt when you continue through its sign-in flow.
Does the text to video AI generator include sound?
Yes. Native audio is generated with the video. W3.0 models always generate audio; Seedance 2.5 has audio enabled by default in the creator. Describe sound effects and ambience in the same prompt. Audio has no separate credit charge.
Can I generate 1080P, 2K, or 4K video?
Yes, with an active Basic, Pro, or Max membership. W3.0, W3.0 Prime, and Seedance 2.5 support up to 1080P. W3.0 Pro and Prime Pro support 1080P, 2K, and 4K and are member-only models. Credit packs alone allow 480P and 720P on non-Pro models.
How long can my video be, and which aspect ratios can I choose?
W3.0 models support 2 to 30 seconds; Seedance 2.5 supports 4 to 30 seconds. For text to video, choose 16:9, 9:16, 1:1, 4:3, or 3:4. The creator shows the available settings for your selected model.
How can I get realistic AI video generator results?
Start with one scene and a clear action. Describe natural lighting, materials, and one camera movement, such as a slow dolly-in. Keep sound direction consistent with the scene. Results vary by prompt, model, and settings, so you may need to refine the prompt and generate again.
Is Wan3 text to video AI free?
Generation requires purchased credits; new accounts do not receive free signup credits. You can buy a credit pack or a membership. The total depends on duration, resolution, and model, and is shown before generating. Membership is required for Pro models and 1080P or higher output.
Can I download my video, and what happens if generation fails?
Completed videos can be downloaded as MP4 files with their generated audio. If a generation fails, its reserved credits are returned automatically.