Generate AI videos from text, images, or reference videos. Wan 3.0 supports output up to 1080p and clips up to 30 seconds when no reference video is uploaded.
Choose a version by input type, output duration, and resolution. The options below describe the features available on this site.
Generate from text, images, or visual references with Standard and Prime options. When using reference video, the combined input and output limit is 30 seconds.
Generate from text or an image, with 2–15-second clips and 720P / 1080P output.
Generate from text or an image, with 5 / 10 / 15-second clips and 720P / 1080P output.
Generate from text or an image, with 5 / 10-second clips and 720P / 1080P output.
Choose text, image, or reference generation, then set duration, aspect ratio, and resolution. See the credit cost before generating, then play and download the result in My videos.
Without a reference video, Wan 3.0 can generate up to 30 seconds. With reference videos, input and output duration together must stay within 30 seconds; reference inputs can total up to 15 seconds.
Upload reference images or short videos to guide the appearance of people, objects, and settings. Clear references and specific prompts help communicate your intent; identity and details can still vary between results.
Describe camera movement such as a push-in, pullback, tracking shot, or orbit in your prompt. The resulting movement depends on the prompt, source material, and generation.
Compare the features currently connected on this site. Available settings and credit costs are shown in the generator; speed and visual results vary by task.
| Feature / Spec | Wan 2.5 | Wan 2.6 | Wan 2.7 | Wan 3.0 ✦ |
|---|---|---|---|---|
| Output resolution | 720P / 1080P | 720P / 1080P | 720P / 1080P | 480P / 720P / 1080P |
| Output duration | 5 / 10s | 5 / 10 / 15s | 2–15s | 2–30s; reference video reduces maximum |
| Text to video | Yes | Yes | Yes | Yes |
| Image to video | Yes | Yes | Yes | Yes |
| Reference image mode | Not available | Not available | Not available | Yes |
| Reference video uploads | Not available | Not available | Not available | Yes |
| First and last frame inputs | Not available | Not available | Not available | Yes |
| Audio or document uploads | Not available | Not available | Not available | Not available |
Wan 3.0 helps creators, marketing teams, and production houses turn concepts into polished video deliverables faster.
Short-Form Content Creation
Use Wan 3.0 to turn text prompts or images into short-form videos for TikTok, YouTube Shorts, Reels, and other social platforms. Strong consistency helps creators produce more content effortlessly.
Brand Marketing & Advertising
Wan 3.0 makes it easier for marketing teams to generate product videos, commercial teasers, and branded visual campaigns online with high-resolution output and realistic movement.
Film & Pre-Visualization
Wan 3.0 helps directors, filmmakers, and creative teams turn scripts, shot ideas, and scene concepts into visual references. Use it for storyboards, pitch decks, and early camera planning.
Choose your monthly creative allowance. Renews monthly; stop future renewal anytime.
1,000 credits / billing period
Billed monthly, renews automatically
2,400 credits / billing period
Billed monthly, renews automatically
5,000 credits / billing period
Billed monthly, renews automatically
Credit usage varies by model, duration and resolution. Confirmed payments grant the current allowance, valid until the billing period ends, with no rollover. Permanent credits are unchanged.
All purchases are charged in US dollars (USD) · Stripe-hosted checkout
WanAI is an independent AI video creation website offering Wan models. Generate short clips from text, images, and reference videos, then view and download results in My videos.
Wan 3.0 can be used for product videos, ads, social content, cinematic scenes, explainers, concept videos, and reference-driven creative projects.
Without a reference video, Wan 3.0 can generate up to 30 seconds. With reference videos, input and output duration together must stay within 30 seconds; reference inputs can total up to 15 seconds.
Yes. Wan 3.0 supports video output up to 1080p, making it suitable for product content, advertising, websites, presentations, and other higher-resolution workflows.
The current inputs are text prompts, JPG / PNG / WebP images, and MP4 / MOV reference videos. Images are limited to 10 MiB each. Reference videos are limited to 100 MiB each, 1–15 seconds per clip, and 15 seconds total. Audio and document uploads are not currently available on this site.
Yes. Upload an image and describe the motion, camera direction, environment, and ending you want. Wan 3.0 can use the image as visual guidance for the generated video.
Audio generation is enabled for current Wan 3.0 requests. You can describe ambience or sound effects in your prompt, but the model determines the result. Separate audio editing and audio uploads are not available on this site.
Choose text, an image, or a reference video, review the credit quote, and start generating. Play and download completed results in My videos.
Up to 1080p · Upfront credit quotes · Video history