Wan-2.5 AI Video Generator | Text & Image to Video Guide
Explore Wan text-to-video and image-to-video workflows, historical Wan 2.5 capabilities, and current Wan 3.0 model resources.
WanVideoMaker is independent and is not affiliated with, endorsed by, or sponsored by Alibaba Group.
WanVideoMaker is an independent Wan information and tutorial resource.
AI Video Generation
Choose your preferred method to create professional videos with our advanced AI technology
Text-to-video workflows start from a prompt and typically combine subject, action, scene, camera direction, visual style, aspect ratio, resolution, and duration.
Wan model
Up to 1080p
Multiple durations
Multiple aspect ratios
WanVideoMaker documents Wan workflows and model capabilities. This section is a workflow reference, not a live generator.
Wan video example
AI Powered
Wan Video
Text to Video
Prompt based
Image to Video
Image animation
Video Quality
Up to 1080p
Wan 2.5 in the Wan video model timeline
Wan 2.5 Preview is now a historical Wan generation. It remains useful for understanding the evolution of Wan text-to-video and image-to-video workflows, but WanVideoMaker no longer presents it as a current paid SaaS product.
Alibaba Cloud documented Wan 2.5 Preview for short-form video generation with text-to-video and image-to-video model variants. The verified model-level specifications below are kept separate from the old WanVideoMaker credit system and promotions.
What is still useful from Wan 2.5 today
The model generation has changed, but the core creative workflow is still useful: describe the subject, motion, scene, camera, lighting and style clearly; then change one variable at a time when iterating.
Prompt structure
Separate subject, action, environment, camera movement and visual style instead of writing one vague sentence.
Image-to-video control
Use a strong source image and describe only the motion or camera change you actually want.
Iteration
Compare outputs after one prompt change at a time so you can tell which instruction improved or harmed the result.
Wan text-to-video and image-to-video workflows
WanVideoMaker keeps these workflows as educational references and focuses on model documentation, pricing references, tutorials and practical comparisons.
- 1
Text to video
Start with a clear subject and action, add the scene and camera direction, then refine style and lighting.
- 2
Image to video
Choose an image with a stable subject and composition, then describe motion, camera movement and what must remain unchanged.
- 3
Evaluate the result
Check subject consistency, motion physics, camera compliance, temporal artifacts and whether unwanted objects changed.
For current Wan generation, use Wan 3.0
Wan has moved beyond Wan 2.5. Current Wan 3.0 documentation covers longer generation, native audio, first/last-frame control and multimodal references. Use the current guide and pricing reference before starting new work.
WanVideoMaker FAQ
Does WanVideoMaker still generate videos directly?
No. WanVideoMaker is now an independent Wan information resource and no longer provides a live generation service.
Is Wan 2.5 still the current Wan model?
No. Wan 2.5 is a historical generation. For current capabilities and pricing, use the Wan 3.0 pages on this site.
Does WanVideoMaker still promise free credits, watermark-free output or commercial rights?
No. Those were old product-level claims and promotions and are no longer presented as current promises. Access terms, pricing and usage rights depend on the current provider and product you use.
Why keep Wan 2.5 content at all?
Older tutorials still have educational and search value when they are clearly labeled as historical. We preserve useful workflow guidance and correct outdated product claims instead of pretending old information is current.
Explore the current Wan model family
Use the Wan 3.0 guide and pricing reference to compare current capabilities, workflows, and provider costs.