Wan 2.7 AI Video Generator
Use Wan 2.7 on Epochal to generate AI videos from prompts or reference images with 720p or 1080p output, 5 to 15 second clips, and prompt expansion.
What is the Wan 2.7 AI Video Generator?
Wan 2.7 is the Wan video model exposed on this page. Alibaba Cloud's official Wan video docs describe a multimodal family that supports text-to-video up to 15 seconds at 1080P with prompt rewriting, while the newer wan2.7 image-to-video API adds first-frame, first-and-last-frame, and continuation workflows. On Epochal, the current page focuses that family into a practical short-form setup for both prompt-led and reference-led video creation.
Key Capabilities
Wan officially supports both text-to-video and image-to-video
Alibaba Cloud positions Wan as a video family that spans text-to-video and image-to-video rather than a single prompt-only workflow.
15-second video with prompt rewriting
Official Wan text-to-video docs highlight support for clips up to 15 seconds, 1080P output, and prompt rewriting to improve short or rough prompts.
New wan2.7 image-to-video tasks
The official wan2.7 image-to-video API supports first-frame generation, first-and-last-frame interpolation, and continuation from an existing clip.
Compact controls on Epochal
On Epochal, the current page exposes 720p or 1080p output, 5 to 15 second duration, prompt expansion, negative prompt, and up to 2 reference images for image-led runs.
How to Use Wan 2.7 Video
Write the subject, motion, camera direction, and scene mood you want, or upload up to 2 reference images when the clip should follow a specific character, product, or composition.
For text-to-video, choose 16:9, 9:16, 1:1, 4:3, or 3:4. Then pick 720p or 1080p and set a duration from 5 to 15 seconds before you run the shot.
Use negative prompt to rule out unwanted motion or visual traits, and turn on prompt expansion when a short prompt needs a stronger rewrite before generation.
Review motion, framing, pacing, and subject consistency, then tighten the prompt or swap references for the next pass if the shot still needs refinement.
Pricing & Credits
Each generation with Wan 2.7 consumes credits inside Epochal.
Processing time varies with queue state, selected resolution, chosen duration, workflow type, and prompt complexity.
Use the live workflow cost shown on the page as the current credit reference. On Epochal, Wan 2.7 cost changes with resolution, duration, and workflow type.
Use Cases
Prompt-led short explainers and concept videos
Use Wan 2.7 when a written idea needs to become a clean short-form motion draft without starting from source footage.
Reference-led motion from key visuals
Use the image-led workflow when a product still, character image, or storyboard frame should anchor the motion direction.
Aspect ratio and resolution comparisons
It works well when the same concept needs to be tested across portrait, square, and landscape outputs at 720p and 1080p.
Prompt refinement workflows
It is useful when prompt expansion and negative prompt should work together to open up or narrow the first video result.
Output & Quality
Wan 2.7 is strongest when you need short controlled clips with clear output settings, prompt refinement, and the option to move between prompt-led and reference-led generation inside one model family.
- - Short prompt-led videos with explicit duration and resolution control
- - Reference-led motion tests built from key frames or product stills
- - Teams comparing portrait, square, and landscape output on the same concept
- - Workflows that benefit from prompt expansion and negative prompt together
- - Best suited to short clips rather than continuity-heavy long-form storytelling.
- - Results are strongest when the prompt or references already lock the subject and motion direction clearly.
- - If your task depends on explicit first-and-last-frame interpolation or continuation from an existing clip, the current simplified page may need more iteration.
FAQ
What does Wan officially support for video generation?
Alibaba Cloud's official Wan docs describe text-to-video with prompt rewriting, up to 15-second output, and 1080P support. The newer wan2.7 image-to-video API adds first-frame generation, first-and-last-frame generation, and continuation from an existing clip.
What controls does the Wan 2.7 AI video generator expose on Epochal?
On the current page, Wan 2.7 exposes 720p or 1080p output and 5 to 15 second duration for both video workflows. Text-to-video adds aspect ratio, while both workflows expose negative prompt and prompt expansion, and image-to-video supports up to 2 reference images.
Does this page support both text-to-video and image-to-video?
Yes. The current Wan 2.7 page covers both prompt-led video generation and reference-led video generation inside the same model family.
How long can Wan 2.7 AI video generator clips be on Epochal?
The current page offers a duration range from 5 to 15 seconds, with 720p or 1080p output. That fits the short-form positioning of the workflow exposed here.
When should I turn on prompt expansion for the Wan 2.7 AI video generator?
Turn prompt expansion on when the prompt is short, rough, or missing camera and motion detail. It is most useful when you want the model to rewrite a basic idea into a fuller video instruction before rendering.
Related Models
Related Tools
Latest Blog Articles
Keep reading the newest posts on model capabilities, workflow tips, and creative practice.
Best Image to Video AI Tools in 2026: Which One Preserves Your Frame Best?
A practical guide to the best image to video AI tools in 2026, comparing Kling 3.0, Veo 3.1, Seedance 2.0, Wan 2.7, and Grok Imagine Video for frame preservation, motion quality, speed, and workflow fit.
Best AI Video Generator in 2026: Veo 3.1, Kling 3.0, Seedance 2.0 and More, Tested
A practical comparison of the best AI video generators available in 2026, covering output quality, audio generation, prompt control, speed, and which model fits each workflow.
Veo 3.1 vs Seedance 2.0: Which One Fits Your Content Workflow?
If you are comparing Veo 3.1 and Seedance 2.0, this guide breaks down where each model fits best across quality, control, output speed, and commercial use.