Wan 2.7 combines four video workflows with Standard and Pro image generation and editing. Create a video from text, one or two frames or image/video references, or edit existing footage with written instructions and an optional target image. For a still image, describe the composition or provide one to four editing references, choose a preset or custom pixel canvas and request one to four outputs. Select the intended result first, then match the source material and output controls to the task’s per-second or per-image estimate.
Wan 2.7 is developed by Alibaba. Vidgo provides access to the model. Model names and trademarks belong to their respective owners.For video, choose a written scene, frame animation, reference-led generation or source-footage editing. For images, select Standard or Pro and choose text generation or editing with one to four sources. Prepare the intended subject, action or composition, identify what to retain and review the exact task’s duration, canvas and billing unit.
Create 5–15 second videos from text prompts with Thinking Mode, natural motion, and optional synchronized audio.
Animate still images into 2–15 second videos with optional first and last frames while preserving subject details.
Turn reference images or videos into cinematic clips with consistent characters and multi-shot storytelling.
Restyle source footage with text instructions and an optional reference image while retaining camera motion and pacing.
Generate 1K or 2K images from text with Thinking Mode, bilingual typography, and palette-guided colors.
Edit and blend 1–4 reference images with text instructions while preserving subject identity and lighting.
Create native 4K images from text with detailed textures, typography, and complex composition control.
Retouch and combine 1–4 reference images with precise local edits and native alpha transparency.
Video image input uses one or two ordered frames. Reference generation accepts images, videos or both; explain the identity, object or environment each source supplies. Video editing uses one video_url and an optional reference_image_url. Image editing uses one to four ordered image_urls. Assign those images to people, objects, a setting or material and describe one integrated result, then select the video settings or image canvas and n.
| Model / task | Input and workflow |
|---|---|
| Wan 2.7 Text To Video | Create 5–15 second videos from text prompts with Thinking Mode, natural motion, and optional synchronized audio. |
| Wan 2.7 Image To Video | Animate still images into 2–15 second videos with optional first and last frames while preserving subject details. |
| Wan 2.7 Reference To Video | Turn reference images or videos into cinematic clips with consistent characters and multi-shot storytelling. |
| Wan 2.7 Video Edit | Restyle source footage with text instructions and an optional reference image while retaining camera motion and pacing. |
| Wan 2.7 Image Standard Text to Image | Generate 1K or 2K images from text with Thinking Mode, bilingual typography, and palette-guided colors. |
| Wan 2.7 Image Standard Edit | Edit and blend 1–4 reference images with text instructions while preserving subject identity and lighting. |
| Wan 2.7 Image Pro Text to Image | Create native 4K images from text with detailed textures, typography, and complex composition control. |
| Wan 2.7 Image Pro Edit | Retouch and combine 1–4 reference images with precise local edits and native alpha transparency. |
Existing records show a sea-arch shot, a fox moving between two frames, an explorer with a referenced lantern, a bilingual timetable and a four-source coastal image. The showcase also includes a video coat edit. Inspect the actual inputs and outputs to understand how the video and image tasks use their different controls.
Write the camera path, water motion and structure to retain. The sea-arch request asks for a low forward push through a basalt cave with a turquoise swell and surf ambience. Inspect geometry, water movement and the requested light in the actual clip.
Provide the beginning and ending frames and explain the transition. The fox request asks for an uncurling motion into a standing pose while preserving fur, face, tail and snowy woodland. Compare both inputs with the actual body mechanics and final state.
Assign one image to the explorer and another to the brass lantern and enable multi_shots. The cave record asks for a careful step, then a closer shot as the lantern illuminates minerals. Inspect appearance, prop shape and the requested sequence against the references.
Write the heading, each row and the physical surface. The Standard timetable record asks for five Chinese-English lines on a front-facing railway board. Inspect wording, alignment and enamel texture before using the image as an editorial or narrative draft.
Assign the sources to a man, a pelican, a boardwalk and a knitted hat. The Pro edit asks for the man and bird to share the coastal setting with matched light and scale. Compare all four inputs with the output for identity, contact and the requested accessory.
The video case supplies walking footage and a teal-coat reference, then asks to change the coat while retaining the woman, movement, street and timing. The Pro image case supplies a man, a pelican, a seaside boardwalk and a mustard beanie, then asks for one integrated photograph with matched scale and light. Read the full prompts and compare every source with its own video or image result.
Edit only the walking woman's coat to match the teal wool coat in the reference image. Preserve her face, hair, body, stride, coat cut, trousers, shoes, camera motion, street, lighting and timing exactly. The coat stays consistently teal in every frame. Do not alter any other object. No text, lettering, captions, logos, watermarks, brands, advertising or product packaging.
Create one realistic coastal observation photograph from the four reference images. Image 1 supplies the exact adult man identity, face, teal jacket and dark trousers. Image 2 supplies the single brown pelican with its distinctive long bill and plumage. Image 3 supplies the empty seaside wooden boardwalk and ocean setting. Image 4 supplies the mustard knitted beanie: put that same beanie on the man. Place the man standing to the left, hands relaxed, and the pelican standing naturally on the boardwalk railing to the right, several feet away. Keep both heads unobstructed and sharp, natural relative scale and anatomy. Harmonize all elements to the overcast light of image 3 with coherent contact shadows. One man, one pelican, one beanie, no duplicates, no text. No advertising, brands, logos, watermark, or foxes.
Prepare a written brief, source frames, reference clips or editing images and choose the required video or image task. Describe the intended change and the details to retain, then configure its actual output settings. Reviewed results can support storyboards, character motion, video wardrobe revisions, bilingual scene graphics, multi-source photographs and alternative visual styles.
Describe a scene, physical motion, camera and sound, then choose the text-video duration and resolution. Review the clip for a narrative sequence or environmental storyboard.
Provide frames or image/video references, assign subject and prop roles and describe the action sequence. Inspect the generated movement and identity for a character or story-development proposal.
Supply the source clip and an optional target image, then identify the garment or surface and retained motion. Compare the revised video with both sources for a wardrobe or material-development concept.
Write the exact Chinese and English wording, line order and surface for an image request. Check spelling and placement in the output before using the reviewed result in an editorial or narrative layout.
Prepare one to four ordered sources for identity, objects, destination and accessories. Describe a coherent composition and inspect scale, light and contact for a scene proposal or illustration draft.
Supply the original image and describe the material, style or object change, then select a canvas and output count. Compare the results with the source for an editorial concept, restoration study or visual-development series.
Video tasks use the selected 720p/1080p rate multiplied by output seconds. Automatic editing uses duration 0: the service reads a 2–10-second source, rounds its duration down to whole seconds and uses that same length for generation and billing. Explicit editing selects 2–10 output seconds. Image tasks use the Standard or Pro per-output-image rate multiplied by n; the current image tier includes the selected canvas and reference processing. The dynamic table retains each task’s current unit and starting rate; configure the actual task for the complete estimate.
| Model | Provider | Workflow | Input material | Output controls | Starting rate |
|---|---|---|---|---|---|
| Wan 2.7 Text To Video | Alibaba | Text to Video | Text prompt; optional audio | 1080p | $0.06 / output second |
| Wan 2.7 Image To Video | Alibaba | Image to Video | 1–2 starting/ending frames | 1080p | $0.06 / output second |
| Wan 2.7 Reference To Video | Alibaba | Reference to Video | Image / video references | 1080p | $0.06 / output second |
| Wan 2.7 Video Edit | Alibaba | Video Editing | 1 video; optional target image | 1080p | $0.06 / output second |
| Wan 2.7 Image Standard Text to Image | Alibaba | Text to Image | Text prompt | Preset / custom pixels | $0.021 / output image |
| Wan 2.7 Image Standard Edit | Alibaba | Image Editing | 1–4 reference images | Preset / custom pixels | $0.021 / output image |
| Wan 2.7 Image Pro Text to Image | Alibaba | Text to Image | Text prompt | Preset / custom pixels | $0.0525 / output image |
| Wan 2.7 Image Pro Edit | Alibaba | Image Editing | 1–4 reference images | Preset / custom pixels | $0.0525 / output image |
Test your creative brief online, then bring the verified request into your application.
Open the selected task from the model list above. Write the motion, scene or image-edit instructions and provide its exact source combination. Configure the supported video duration or image canvas and count, then review the full estimate. Compare the result with every source for identity, structure, material, motion and requested changes.
Choose a task, inspect its settings and cost, and follow the result from an online test to an authenticated API request.
Choose video or image from the model list. For video, match text, frames, references or source editing to the supported duration; for images, select Standard/Pro and text or one-to-four-image editing. Prepare each source’s role and choose the task’s resolution, canvas, image count and other supported controls.
Open the selected task from the model list above. Write the motion, scene or image-edit instructions and provide its exact source combination. Configure the supported video duration or image canvas and count, then review the full estimate. Compare the result with every source for identity, structure, material, motion and requested changes.
For video, use the selected output seconds and resolution; let automatic editing finish reading the source duration for its whole-second estimate. For images, choose Standard or Pro and the actual n. Compare complete amounts and planned test counts using each task’s own billing unit.
Read the actual prompt and inspect every source image or video alongside its generated result. Use the recorded examples to understand how to describe input roles, action progression and the details to preserve in your own request.
Retain the task ID returned by the API to query status and retrieve completed file URLs. Use your online generation history to review earlier results and the creative direction you tested.
Use the documentation linked from each task for its exact model identifier, accepted inputs, authentication and response format. Integrate the same task you tested online so its media and settings match the request.
Wan 2.7 is one Alibaba family on Vidgo covering four video tasks and four Standard/Pro image tasks. Video supports text, frames, image/video references and editing at 720p/1080p. Images support text generation and one-to-four-reference editing with preset/custom pixel canvases and one to four outputs.
Explore model introductions, API guides and platform comparisons to understand input preparation, request integration and cost planning for your project.
