Wan 2.7
Wan 2.7 is Alibaba Tongyi Lab's video model, released in the first week of April 2026 as a suite of four hosted models: text-to-video, image-to-video, reference-to-video with voice cloning, and instruction-based editing. It adds a thinking mode and stronger character consistency to the Wan line and keeps native audio. Despite the family's open-weight reputation, nothing after Wan 2.2 has been published: 2.5, 2.6, 2.7 and 3.0 are API-only, and the Wan-AI organisation on Hugging Face still tops out at the 2.2 video weights. On fal it costs $0.10 a second at 720p or $0.15 at 1080p. Wan 3.0 followed in August and has its own page.
Anvisha Pai, Co-founder & CEO, Voyager
Alibaba · released
›What we asked for
A self-portrait of Wan 2.7, Alibaba's open-weight video model: an ink-brush calligrapher's hand writes a single flowing character on rice paper, the ink blooming, the camera pulling back to reveal a studio full of hanging scrolls, quiet brush sounds.
fal-ai/wan/v2.7/text-to-video
aspect_ratio: 16:9, resolution: 720p, duration: 5
returned 1280 by 720 pixels
At a glance
- Price
- $0.10 a second at 720p, $0.15 at 1080p on fal. Our five-second clips cost $0.50.
- Weights
- None to download. The last open Wan video weights are 2.2; 2.7 is hosted only.
- Length
- 2 to 15 seconds.
- Audio
- Native, always on, and an audio_url input to drive the clip with your own track.
- Shapes
- 16:9, 9:16, 1:1, 4:3 and 3:4.
Other Wan models: Wan 3.0. Compare all 2.
Anvisha's take
Wan 2.7 is a solid mid-price hosted model, not the open-weight option its family name suggests: the newest Wan video weights you can download are 2.2. On the clips, the harbour is a real aerial, the image-to-video clip put the dripper on a carafe and poured properly, and the calligraphy self-portrait is one of the better heroes. It also tripped over the fixture set in two ways nobody else did: fal's content checker refused the festival title card outright, and the physics clip grew the ice cubes to the size of the glass. $0.50 a clip; Wan 3.0 is the same money for twice the length.
Wan 2.7 real output: the same 9 clips every model gets
Every video model on this site gets the same 9 clips, so you can compare like with like. These are the clips Wan 2.7 returned, one try each, nothing re-rolled or retouched. Each tile shows one frame of the clip until you press play. Generated on . The 8 clips took 41 to 134 seconds each and cost $4.00 in total.
A clean product shot that holds its shape through a full slow rotation.
›What we asked for
A matte black ceramic pour-over coffee dripper rotating slowly on a white turntable, soft studio light, seamless white background, product video, no text.
fal-ai/wan/v2.7/text-to-video
aspect_ratio: 16:9, resolution: 720p, duration: 5
returned 1280 by 720 pixels
A face, lip sync and a spoken line, where the model makes audio.
›What we asked for
A woman in her 40s with short grey hair stands in a bright kitchen, looks at the camera and says 'Good morning, the coffee is ready', natural window light, handheld feel, she smiles at the end.
fal-ai/wan/v2.7/text-to-video
aspect_ratio: 16:9, resolution: 720p, duration: 5
returned 1280 by 720 pixels
Set text that stays spelled correctly while it moves.
›What we asked for
A title card reading 'MODA SUMMER SESSIONS' animates in over slow-motion footage of a festival crowd at dusk, then a smaller line 'Friday 12 June' fades in beneath it. Bold condensed type, tangerine on navy.
fal-ai/wan/v2.7/text-to-video
aspect_ratio: 16:9, resolution: 720p, duration: 5
A steady drone push-in with stable geometry.
›What we asked for
Drone shot pushing slowly forward over a coastal town at sunrise, terracotta roofs, fishing boats in the harbour, long shadows, cinematic and steady.
fal-ai/wan/v2.7/text-to-video
aspect_ratio: 16:9, resolution: 720p, duration: 5
returned 1280 by 720 pixels
A spill in slow motion: does water behave like water.
›What we asked for
A glass of water tips over on a wooden table and spills in slow motion, water and three ice cubes sliding across the wood, macro lens, morning light.
fal-ai/wan/v2.7/text-to-video
aspect_ratio: 16:9, resolution: 720p, duration: 5
returned 1280 by 720 pixels
Hands, a kettle and a steady pour without extra fingers or teleporting objects.
›What we asked for
Close-up of hands pouring hot water from a gooseneck kettle in slow circles into a pour-over coffee dripper, steam rising, shallow depth of field, warm cafe light.
fal-ai/wan/v2.7/text-to-video
aspect_ratio: 16:9, resolution: 720p, duration: 5
returned 1280 by 720 pixels
Many people and objects moving at once without merging.
›What we asked for
A crowded Saturday farmers market seen from slightly above, shoppers with tote bags, a dog on a leash, a busker with a guitar, bunting swaying, late morning light, gentle camera drift.
fal-ai/wan/v2.7/text-to-video
aspect_ratio: 16:9, resolution: 720p, duration: 5
returned 1280 by 720 pixels
A named style held across every frame rather than drifting to photoreal.
›What we asked for
A lighthouse on a rocky cliff in a storm, in the style of a Japanese ukiyo-e woodblock print, waves crashing in flat colour with visible wood grain, animated like moving paper cutouts.
fal-ai/wan/v2.7/text-to-video
aspect_ratio: 16:9, resolution: 720p, duration: 5
returned 1280 by 720 pixels
Bring a still product photo to life without changing the product.
›What we asked for
The coffee dripper slowly fills as coffee drips through it, steam rising, the camera drifting gently around it, the dripper itself unchanged. (starting from The Nano Banana Pro product shot from the image fixture set.)
fal-ai/wan/v2.7/image-to-video
resolution: 720p, duration: 5
returned 960 by 960 pixels
How the cost under each picture was worked out: fal's Wan 2.7 rate of $0.10 per second at 720p (fal model page, read 2026-09-09) for the 5 seconds requested; fal does not return the charge.
Wan 2.7 pricing
| Where | Unit | Price | Source |
|---|---|---|---|
| fal | Second at 720p $0.15 a second at 1080p; the reference-to-video endpoint is $0.10 at both. | $0.10 | fal model page, read |
| Alibaba Model Studio | Second at 720p Alibaba Cloud lists $0.10 at 720p and $0.15 at 1080p for the hosted model. | $0.10 | Coverage of Wan 2.6 and 2.7 pricing, read |
| fal | Five-second 720p clip | $0.50 | fal model page, read |
Is Wan 2.7 free?
Alibaba's Wan app gives daily free credits. There are no Wan 2.7 weights to run yourself, and the hosted endpoints charge per second with no free tier.
- •Wan app: daily free credits from Alibaba.
- •Open weights: none for 2.5, 2.6, 2.7 or 3.0; the Wan-AI organisation on Hugging Face stops at the 2.2 video models and later control models such as Animate.
- •fal, Alibaba Model Studio and other hosts: per second.
Wan-AI on Hugging Face, newest video weights are Wan 2.2, read
Where to use Wan 2.7
- Wan app · Alibaba's consumer app with daily credits.
- Alibaba Model Studio · Alibaba Cloud's hosted API.
- fal · Per-second by resolution; text, image and reference endpoints.
- Voyager · Runs inside Voyager, the agent for creative work. Private preview by waitlist.
Voyager
Run Wan 2.7 inside Voyager
Voyager is an agent for creative work that runs every model in this reference, Wan 2.7 included. Brief it, and it picks the model, makes the thing and hands back something you can edit. Create an account to get started.
What we noticed running Wan 2.7
- •Refused: fal's content checker returned a 422 on the prompt, a festival title card over a crowd. No other model flagged it. See the animated title card clip.
- •The glass tips and the ice comes out, then the cubes swell to nearly the size of the glass. Object scale drifted inside the clip. See the physics clip.
- •Opened wide on the turntable and cut to a close-up of the dripper on the same shot; the model added an edit to a one-shot prompt. See the product turntable clip.
- •Followed the square starting image, so the clip came back 960 by 960; the image endpoint has no aspect ratio field. See the image to video clip.
Wan 2.7 limits and API parameters
| Limit | Value |
|---|---|
| Length | 2 to 15 seconds. fal API reference, read |
| Resolution | 720p or 1080p. fal API reference, read |
| Audio | Always on; not switchable off on the hosted endpoints. fal API reference, read |
›Every setting, for developers
fal (fal-ai/wan/v2.7/text-to-video)
fal API reference, read
| Parameter | Type | Values | Default |
|---|---|---|---|
| prompt | string | Up to 5,000 characters Required. | |
| duration | integer | 2 to 15 seconds | 5 |
| resolution | enum | 720p, 1080p | 1080p |
| aspect_ratio | enum | 16:9, 9:16, 1:1, 4:3, 3:4 | 16:9 |
| audio_url | URL | A 3 to 30 second WAV or MP3 to drive the clip | |
| negative_prompt | string | Up to 500 characters | |
| enable_prompt_expansion | boolean | true, false | true |
| seed | integer | 0 to 2147483647 | |
| image_url | URL | Starting frame Image-to-video endpoint, which has no aspect ratio field. |
What Wan 2.7 will not make
Wan's hosted models run behind fal's safety checker, on by default and only switchable off with account authorisation, and under Alibaba Cloud's Model Studio terms. The practical boundaries are sexual content, minors, violence and deceptive depictions of real people. In our run the checker refused a festival title card on the 2.7 endpoint.
fal API reference, read
Alternatives to Wan 2.7
Frequently asked questions
What is Wan 2.7?
Wan 2.7 is Alibaba Tongyi Lab's hosted video model suite from early April 2026, with native audio, a thinking mode and voice cloning on the reference endpoint. It is on fal at $0.10 a second at 720p.
Is Wan 2.7 open source?
No. Alibaba has published no weights after Wan 2.2; the Wan-AI organisation on Hugging Face stops there for video models, read on 9 September 2026. Wan 2.5, 2.6, 2.7 and 3.0 are API-only.
Is Wan free?
The Wan app gives daily credits, and the Wan 2.2 weights are free to run yourself under Apache 2.0. Wan 2.7 and 3.0 are hosted only and charge per second.
Is Wan 3.0 out?
Yes: public beta from Alibaba since 6 August 2026 and on fal since late August. It has its own page here with the same nine clips.
How this page is made
Prices, limits and settings are read from the linked vendor and host pages on the dates shown. Every picture was generated by us through fal with the request shown under it, and the original files are kept unedited. A generation that failed is shown as a failure. Model and vendor names are used to identify the products; no vendor imagery appears on this page.
