AI Video Generator
Alibaba's Wan 2.7 makes the frame and then moves it: stills, edits and clips with sound from one model family.
Output
Four generations of Alibaba's Wan video model run on this site, and a newer one is not always dearer. A five second 720p clip costs 126 credits on Wan 2.6 or 2.7, 150 on Wan 2.5 and 180 on Wan 3.0, so the right pick depends on the length, the size and whether you start from a picture.
September 2025, with a low-cost 480p tier
Five or ten second clips at 480p, 720p or 1080p, from text or from a still. At 480p a five second clip costs 75 credits, the lowest price for a five second Wan clip written from text. From 720p up, Wan 2.6 and 2.7 cost less and do more.
Open the Wan 2.5 pageDecember 2025, with the low-cost Flash tier
Text and image to video in 5, 10 or 15 second clips at 720p or 1080p. Its Flash tier animates a still at 720p for 75 credits with sound or 38 without, the lowest price for a five second Wan clip from a picture.
Open the Wan 2.6 pageApril 2026, stills and clips from one model
Text and image to video at any whole second from 2 to 15, at 720p or 1080p, for the same price as Wan 2.6. The same page also makes stills: text to image and image to image at 15 credits each.
The newest, up to 30 seconds
Text, image and reference to video, where the reference mode takes images and a motion clip. Any length from 2 to 30 seconds, with a switch for sound. A five second clip is 90 credits at 480p; at 720p and 1080p it costs more than Wan 2.7.
Open the Wan 3.0 pageThe falconer was drawn by Wan 2.7's image model, then handed to Wan 2.7 image to video as the first frame. The hawk spreads its wings, leaves the glove and flies low over the heather as the camera pans after it, wings, tail and the bell on its leg intact in flight, with wingbeats and moor wind in the soundtrack.
Start frameThe falconer raises his gloved arm and the hawk spreads its wings and takes off, flying low over the purple heather toward the fog while he watches it go. The camera stays at shoulder height and pans to follow the bird. Sound: heavy wingbeats, wind over the moor, a small bell on the hawk's leg, no music.
A wide shot of a rowboat at dawn as the rod bends, a close-up of the granddaughter fighting it, then her grandfather laughing. The prompt wrote three timed shots and two lines, and the clip cuts on cue, each line in its own voice on the right face.
A small wooden rowboat on a still lake at dawn, mist on the water, pine forest behind. An old fisherman with a white beard and a young woman in a yellow rain jacket sit facing each other. Shot 1 [0-3s]: wide shot, her fishing rod suddenly bends hard toward the water. Shot 2 [3-5s]: close-up of her, she grips the rod with both hands and shouts: "Grandpa, it's pulling!" Shot 3 [5-8s]: close-up of him, he laughs and says: "Hold on, that's the big one." Sound: their two voices, water lapping the hull, the reel clicking, no music.
A rough pencil sketch of an armchair, itself drawn by Wan 2.7's image model, goes in with a description of the finished piece. The edit returns a walnut chair with rust boucle cushions in a 4:3 studio shot, the armrests, tapered legs and three-quarter angle taken straight from the sketch.
Source imageTurn this pencil sketch into a photorealistic studio product photograph of the same armchair. Keep the exact shape, proportions, curved armrests, tapered legs and three-quarter angle from the sketch. Solid walnut frame with a satin oil finish, seat and back cushions in rust orange boucle fabric, on a seamless warm grey studio backdrop with a soft shadow on the floor. Soft key light from the left, 50mm lens, no text, no logos.
Draft the frame as a still, fix it with an edit, then animate it, all on one page.
Write the shots and the lines in order and get a cut scene with voices, up to 15 seconds long.
Turn a concept sketch into a photoreal product shot that keeps its lines and proportions.
How the model works, based on public documentation and sourced evidence.
Wan 2.7 is the latest generation of Alibaba's Wan video and image models, announced on April 7, 2026. The video lineup covers four jobs: text-to-video, image-to-video, reference-to-video and instruction-based video editing. All video variants output 720P or 1080P at 30fps, with clips from 2 to 15 seconds (2 to 10 seconds for editing).
The headline change is that audio is generated together with the picture, not layered on afterward. The model produces speech, music, and ambient sound synced to the scene, or you can supply your own audio file (WAV or MP3, 2 to 30 seconds) to drive the performance. Multi-shot stories are now directed in plain prompt language with per-shot descriptions, and Alibaba says the model keeps up to five characters consistent across shots, each with their own voice, plus dialogue editing that preserves lip sync.
The companion Wan 2.7 image model adds an official thinking mode, on by default for text-to-image, where the model reasons about the prompt before rendering; Alibaba's docs state it improves output quality at the cost of longer generation time. The image model accepts up to 9 reference images, and a separate pro tier renders text-to-image up to 4K.
Two things creators should know about the fine print. Alibaba has not published a technical report or architecture details for 2.7, and unlike Wan 2.1 and 2.2, which shipped open weights under Apache 2.0, no 2.7 weights appear on the official Wan GitHub as of mid 2026, so access is API and platform only. On provenance, the API offers an optional visible "AI Generated" watermark in the corner of outputs, but it defaults to off.
On Vidney, Wan 2.7 runs text to video and image to video from a first frame at any whole second from 2 to 15, at 720p or 1080p, and every clip comes back with sound. Text to video takes five aspect ratios and follows shots written into the prompt; image to video keeps the shape of the picture. The same page runs the standard Wan 2.7 image model for text to image and for image to image with one reference picture. First and last frame, reference to video, video editing and your own soundtrack are not offered here.
Sources: github.com · wan.video
Wan 2.7 is Alibaba's April 2026 Wan release, and on Vidney it is the one page where the same model family makes both the still and the clip. Draft a frame with the image model, edit it from one reference picture, then animate it from that first frame, without leaving the page or the credit balance. Video runs any whole second from 2 to 15 at 720p or 1080p, from text in five aspect ratios or from a picture, and every clip comes back with sound; shots written into the prompt come back as cuts. Next to Wan 3.0 it costs less at 720p and 1080p and has the image model, while Wan 3.0 runs to 30 seconds and adds reference to video.
Create AI video with Wan 2.7 on Vidney in three steps.
Specs side by side with similar models, all runnable on Vidney.
| Wan 2.7 | Vidu Q4 Preview | Grok Imagine Video 1.5 | Grok Imagine Video 1.5 Lite | |
|---|---|---|---|---|
| Task modes | Text to Video · Image to Video · Text to Image · Image to Image | Image to Video · Reference to Video | Text to Video · Image to Video · Reference to Video | Text to Video · Image to Video |
| Duration | 2–15 seconds | 3–16 seconds | 1–15 seconds | 1–15 seconds |
| Resolution | 720p · 1080p | 540p · 720p · 1080p · 2K · 4K | 480p · 720p · 1080p | 480p · 720p · 1080p (upscaled from 720p) |
| Audio | Included, always on | Supported | Included, always on | Included, always on |
| Cost | 15–126 credits per generation | 159 credits per generation | 18 credits per generation | 18 credits per generation |
Recurring themes from public reviews, comparisons and creator write-ups. Both the good and the bad.
Creators consistently praise the directorial control: first and last frame anchoring, multi-shot prompting, and character reference actually behave as instructed, which reviewers call the real shift over earlier versions rather than flashier visuals.
Hands-on reviews highlight the native audio as a genuine workflow win, with lip sync holding up well on short clips at a normal speaking pace and sound designed into the scene instead of bolted on.
Testers also report clearly better prompt adherence and shot-to-shot coherence than previous Wan releases, especially in complex multi-character scenes.
A common complaint is speed: reviewers report around 4 minutes for a 5-second clip, which makes rapid iteration painful compared to faster rivals.
Testers also flag inconsistent generated speech, describing a meaningful share of clips as muddy or unintelligible, and lip sync drifting on clips past 8 to 10 seconds or fast dialogue.
More AI video models you can run on Vidney.
Up to 30 seconds in one pass, sound included: write the scene, animate a start frame, or cast it from reference images.
Alibaba's Wan 2.6 with its Flash tier: the lowest price per second in the Wan family for putting a still in motion.
Alibaba's first Wan with sound built in: speech, effects and music come out with the picture, from a low-cost 480p tier.
Write the scene, the lines and the sound, and Shengshu's Q3 returns all three together, from one second to sixteen.
Every clip comes back with its own sound, dialogue and lip sync included.
Generate AI video with Runway Gen-4 Turbo: text to video and image to video, all from one Vidney workspace.
Wan 2.7 is an AI video model you can run on Vidney. It supports text to video, image to video, text to image, and image to image from a single workspace.
Wan 2.7 costs 15–126 credits per generation. The exact credit cost is shown on the generate button before you run it. That is about $0.45–$3.78 per run at the Starter plan rate ($0.03 per credit).
Wan 2.7 supports the following aspect ratios: 1:1, 4:3, 3:4, 16:9, 9:16.
Wan 2.7 generates clips of 2–15 seconds.
Yes. Wan 2.7 generates video with sound in the same pass: dialogue, effects and room tone. There is no switch to turn it off.
Partly. New accounts get 10 to 30 free credits depending on country, no card required, and Wan 2.7 costs 15–126 credits per generation. Whether your sign-up credits cover a run depends on your balance; the exact cost is always shown on the generate button before you spend a credit.
Yes. The videos you generate with Wan 2.7 on Vidney are yours to use in commercial projects: ads, social posts, client work, and product content. You keep the output; Vidney only handles generation and storage.
Only on the free plan. Free-plan downloads carry a small Vidney watermark; on any paid plan, Wan 2.7 videos download clean, ready to publish or hand to a client. Upgrading also removes the watermark from videos you already made.
The closest Wan 2.7 alternatives on Vidney are Vidu Q4 Preview, Grok Imagine Video 1.5, and Grok Imagine Video 1.5 Lite. They all run in the same workspace from one credit balance, so you can run the same prompt on each and compare the results side by side.
No. Vidney runs Wan 2.7 for you, so there are no API keys to manage and no separate provider account to set up.
