Wan 2.6 creates videos from text, one starting image or one to ten reference videos at 720p or 1080p. Describe the people or objects, action, camera and sounds; enable multi_shots when the brief calls for several shots within the selected clip. Text and image generation offer 5, 10 or 15 seconds, while video-reference generation offers 5 or 10 seconds. Use a source frame to anchor the starting scene, or reference clips to identify the character for a new setting and action.
Wan 2.6 is developed by Alibaba. Vidgo provides access to the model. Model names and trademarks belong to their respective owners.Choose Text to Video when the prompt establishes the full scene. Choose Image to Video when one frame supplies the appearance and environment to animate. Choose Video Reference to Video when source clips establish the subject for a new action or setting. Match the task’s allowed duration and resolution, then describe one continuous shot or the ordered shots you want.
Alibaba's Wan 2.6 video generation family for text-to-video, image-to-video, and video-to-video with multi-shot 1080p output.
Alibaba's Wan 2.6 video generation family for text-to-video, image-to-video, and video-to-video with multi-shot 1080p output.
Alibaba's Wan 2.6 video generation family for text-to-video, image-to-video, and video-to-video with multi-shot 1080p output.
All tasks use prompt, duration and resolution, with optional multi_shots. Image generation adds exactly one image_urls entry. Reference generation adds one to ten video_urls entries; identify the appearance and scene details each clip supplies and describe the new action. For multi-shot work, describe the order, timing and recurring subject in the prompt and set multi_shots to true.
| Model / task | Input and workflow |
|---|---|
| Wan 2.6 Text-to-Video | Alibaba's Wan 2.6 video generation family for text-to-video, image-to-video, and video-to-video with multi-shot 1080p output. |
| Wan 2.6 Image-to-Video | Alibaba's Wan 2.6 video generation family for text-to-video, image-to-video, and video-to-video with multi-shot 1080p output. |
| Wan 2.6 Reference-to-Video | Alibaba's Wan 2.6 video generation family for text-to-video, image-to-video, and video-to-video with multi-shot 1080p output. |
Existing records show an archaeological two-shot scene, an astronaut starting from a frame, a referenced robot in a greenhouse, a paper-theatre animation and a referenced dog on a beach. Inspect each request and actual clip to see how the input and shot settings define the creative goal.
Enable multi_shots and write the stages and timing. The excavation record uses a wide view followed by a close view of the same archaeologist brushing a fragment and speaking. Inspect the cut, appearance, location and requested line against the ten-second brief.
Provide the frame and describe the motion and retained surroundings. The astronaut record asks for a gentle push from a handrail and a sideways drift while equipment stays fixed. Compare the clip with the source for body shape, face, clothing and camera distance.
Provide the subject clip and identify its defining appearance. The robot request retains the yellow torso, turquoise eye and grey feet while placing the subject on a fern-greenhouse path. Inspect the new setting, movement and requested servo sounds against the video reference.
Use the paper-theatre frame and describe the movement of its existing parts. The dragon record asks for accordion wings to unfold and a small head tilt while retaining planted feet and cut-paper surfaces. Inspect shape, material and requested paper sounds in the output.
Supply the animal video and describe the target environment, gait and camera. The dachshund request places the same dog on a beach, trotting beside a low tracking camera. Inspect body proportions, collar, paw contact and footprints against the reference.
The first case provides an astronaut laboratory frame and asks for a handrail push, a gentle sideways drift and a following camera. The second provides a yellow-robot video and asks for the same subject to inspect a fern inside a new greenhouse. Both use five-second, 720p single-shot requests. Read the complete prompts and compare the image or video source with its own generated clip.
Continue naturally from this exact first frame. The astronaut gently pushes away from the fixed handrail with her right hand, then drifts slowly sideways in zero gravity, her body and clothing retaining their original shape. A subtle camera track follows her at the same distance; equipment remains fixed. Keep her face, blue flight suit, handrail and window consistent. A quiet ventilation hum and soft fabric movement. One uninterrupted shot, no speech, cuts, logos or captions.
Use the exact yellow robot from the reference video as the sole character. Place it on a stone path inside a daylight fern greenhouse. Preserve the mustard rectangular torso, single turquoise eye, short arms with two-finger grippers and broad grey feet. In one full-body three-quarter shot, it bends slightly forward to inspect a large fern frond without touching it, then tilts its rectangular upper body a little to one side. Gentle servo sounds and greenhouse ambience. A completely new leafy glasshouse background. No speech, music, text or cuts.
Prepare a written scene, one starting image or source video references. Describe the action, appearance, camera and sound, then choose the supported duration and multi-shot setting. Reviewed outputs can support narrative previsualization, action continuity studies, illustrated-character motion, new environments for referenced subjects and short spoken scene concepts.
Write ordered shots, their timing and the recurring person or object, then enable multi_shots. Review the generated cuts and continuity for a film concept or story sequence.
Provide one starting image and describe contact, movement and fixed surroundings. Inspect the clip against the original for a science, action or character-motion study.
Provide one to ten reference videos and identify the subject each supplies. Describe the destination and activity, then compare identity and motion in the output for a scene-development proposal.
Supply one scene frame and describe movement in the existing materials and shapes. Review the design, motion and requested sounds for a paper-theatre or illustration-animation concept.
Use a reference video to establish the animal and describe a new environment and gait. Inspect anatomy, ground contact and framing for a wildlife or narrative shot proposal.
Describe the speaker, exact line, visible action and sounds within the selected shots. Review the actual speech, timing and recurring appearance before using the clip in a dialogue storyboard.
Wan 2.6’s table lists per-video starting prices using the task’s supported duration and resolution tiers. Text/image generation offer 5/10/15 seconds; video-reference generation offers 5/10 seconds. The task estimate multiplies the requested output seconds by the selected resolution rate. Use your actual duration and 720p/1080p setting to review the complete video cost before submitting.
| Model | Provider | Workflow | Input material | Output specification | Starting rate |
|---|---|---|---|---|---|
| Wan 2.6 Text-to-Video | Alibaba | Text to Video | Text prompt | 1080p | $0.4 / video |
| Wan 2.6 Image-to-Video | Alibaba | Image to Video | Exactly 1 starting image | 1080p | $0.4 / video |
| Wan 2.6 Reference-to-Video | Alibaba | Video Reference to Video | 1–10 reference videos | 1080p | $0.4 / video |
Test your creative brief online, then bring the verified request into your application.
Enter the scene, recurring subject, actions, camera and sounds, then add the selected task’s image or video references. Configure supported duration, resolution and multi_shots and review the full estimate. Compare the generated clip with the sources, including identity, motion, cuts, materials and requested audio.
Choose a task, inspect its settings and cost, and follow the result from an online test to an authenticated API request.
Choose text for a full written scene, one image for a starting composition or one to ten videos for reference subjects. Set 5/10/15 seconds for text/image generation or 5/10 seconds for video references, then choose 720p or 1080p. Enable multi_shots when your prompt describes an ordered sequence of shots.
Enter the scene, recurring subject, actions, camera and sounds, then add the selected task’s image or video references. Configure supported duration, resolution and multi_shots and review the full estimate. Compare the generated clip with the sources, including identity, motion, cuts, materials and requested audio.
Review the full video tier for the actual duration and selected resolution. Use the complete estimate and planned number of tests when comparing text, image and video-reference workflows.
Read the actual prompt or recorded request input and inspect every source image or video alongside its generated clip. 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 2.6 is an Alibaba video family on Vidgo with text, one-image and video-reference generation. It supports 720p/1080p and optional multi_shots. Text/image requests offer 5/10/15 seconds, while reference requests accept one to ten videos and offer 5/10 seconds.
Explore model introductions, API guides and platform comparisons to understand input preparation, request integration and cost planning for your project.
