Wan 2.7
Wan 2.7 is Alibaba's open-weight, uncensored video generation suite — 27B MoE architecture, native audio, 15s clips, and four production-ready modes under Apache 2.0.
Overview
What is Wan 2.7
Wan 2.7 is Alibaba's open-weight video generation suite, released in April 2026. It includes text-to-video, image-to-video, reference-to-video, and video-to-video variants, all under Apache 2.0. With a 27B Mixture-of-Experts architecture, it supports 1080p clips up to 15 seconds, native audio, and instruction-based editing — making it a flexible, production-ready alternative to closed models.
Running it privately on Venice
On Venice, Wan 2.7 runs under an anonymized privacy tier — your prompts are not stored, profiled, or used for training. You retain full sovereignty over your creative workflow, with uncensored, permissionless access to a model family that’s open under Apache 2.0. This means no Big Tech surveillance, no data retention, and no licensing lock-in.
Assessment
Strengths and limitations
- Open under Apache 2.0: fully self-hostable, fine-tunable, and free from licensing restrictions.
- Four distinct video modes: text-to-video, image-to-video, reference-to-video (with subject consistency), and video-to-video editing.
- Native audio generation in the reference-to-video variant, enabling synchronized voice and sound.
- Uncensored and permissionless: ideal for creators needing full control and no content filters.
- Supports up to 15-second clips at 1080p across multiple aspect ratios.
- No end-to-end encryption or TEE protection on Venice — privacy is anonymized but not fully encrypted.
- Text-to-video and image-to-video variants do not support native audio generation.
- Closed-source competitors like Sora 2 and VEO 3.1 still lead in cinematic quality and lip-sync accuracy.
- No support for resolutions beyond 1080p: 4K is not available.
Samples
Sample outputs
Generated on Venice with our standard prompt suite — the same prompts we run through every model of this type, so you can judge it like-for-like.
Slow aerial drone shot gliding over a misty mountain valley at golden hour, sunlight piercing clouds onto a winding river, ancient pine forests on either side, ultra-smooth motion, professional color grading, atmospheric haze, 4K
Calm ocean waves rolling onto a black-sand beach at sunrise, golden light on wet sand, foam dissolving into the shore, a single silhouetted figure at the waterline, smooth continuous forward push, serene cinematic atmosphere
A slow tracking shot through a rain-soaked Tokyo street at night, neon reflecting in puddles, steam rising from a food cart, a person with a translucent umbrella, shallow depth of field, teal-and-orange grade, smooth steady camera
Capabilities
What it supports
- Text to video
- Image to video
- Reference to video
- Native audio generation
Variants
Wan 2.7 model variants
Wan 2.7 runs on Venice as 4 variants of the same underlying model. Pick by what you're starting from: a written prompt, a still image, reference images, or an existing clip. Each variant is its own model id on the API; the generation quality is the same across the family.
| Variant | What it is | Clip lengths | Resolutions | Aspect ratios | Audio | Model ID |
|---|---|---|---|---|---|---|
| Text to Videoflagship | Generate a clip from a written prompt | 5s, 10s, 15s | 1080p, 720p | 16:9, 9:16, 1:1 | wan-2-7-text-to-video | |
| Image to Video | Animate a still image into motion | 5s, 10s, 15s | 1080p, 720p | — | wan-2-7-image-to-video | |
| Reference to Video | Keep a subject consistent using reference images | 5s, 10s | 1080p, 720p | — | wan-2-7-reference-to-video | |
| Video to Video | Edit or restyle an existing clip | Auto | 1080p, 720p | — | wan-2-7-video-to-video |
Capability data comes straight from the Venice model API and refreshes with every catalog ingest. The specs and pricing on this page are captured from the flagship variant; pass the model id of the variant you want to the API.
Wan 2.7 Text to Video
Generate a clip from a written prompt. Supports clips of 5s, 10s, 15s, 1080p, 720p output, 16:9, 9:16, 1:1 aspect ratios.
wan-2-7-text-to-videoWan 2.7 Image to Video
Animate a still image into motion. Supports clips of 5s, 10s, 15s, 1080p, 720p output.
wan-2-7-image-to-videoWan 2.7 Reference to Video
Keep a subject consistent using reference images. Supports clips of 5s, 10s, 1080p, 720p output, with native audio.
wan-2-7-reference-to-videoWan 2.7 Video to Video
Edit or restyle an existing clip. Supports clips of Auto, 1080p, 720p output.
wan-2-7-video-to-videoSpecifications
Datasheet
- Maker
- Alibaba
- Released
- April 7, 2026
- Modality
- Text-to-video, image-to-video, reference-to-video, video-to-video
- Architecture
- 27B Mixture-of-Experts
- Resolutions
- 1080p, 720p
- Clip lengths
- 5s, 10s, 15s
- Mode
- text-to-video
- Aspect ratios
- 16:9, 9:16, 1:1
- Audio
- No
- Privacy on Venice
- Anonymized — prompts not stored
- Available on Venice since
- Apr 2026
- License
- Apache 2.0
API
Call it from your code
Venice exposes this model through the REST API. Queue a generation with the model id.
curl https://api.venice.ai/api/v1/video/queue \
-H "Authorization: Bearer $VENICE_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "wan-2-7-text-to-video",
"prompt": "Aerial drone shot over a misty mountain valley at golden hour"
}'
# Use the returned queue_id with https://api.venice.ai/api/v1/video/retrieve.
# Call /video/complete after downloading if needed.Pricing
What it costs on Venice
Pay per clip on Venice — price scales with resolution and duration (5s–15s), from $0.55.
New Venice accounts include a free daily allowance and 500 welcome credits — no credit card required.
Alternatives
How it compares
| Model | Max resolution | Strongest at | Open weights | Price (Venice) |
|---|---|---|---|---|
| Wan 2.7 | 1080p | Flexible video modes & editing | Yes | from $0.55 |
| Kling O3 Pro | 4K | Cinematic quality | No | from $0.46 |
| Vidu Q3 | 1080p | Balanced quality & speed | No | from $0.27 |
| Wan 2.7 Enhanced | 1080p | Enhanced motion & detail | Yes | from $0.68 |
The most flexible open-weight video model — supports editing, reference inputs, and multiple modes.
Use cases
What it is good for
- 01Generating short-form social content with consistent branding using reference-to-video.
- 02Animating still images into motion for storytelling or marketing with image-to-video.
- 03Editing existing clips via text instructions — e.g., changing scene lighting or adding effects.
- 04Creating uncensored, open-licensed video content for commercial or activist use.
- 05Prototyping video ideas quickly using text-to-video with no data retention.
Prompting
Getting better results
Use the reference-to-video variant with one or more reference images to maintain character or object consistency.
For longer narratives, chain multiple 15-second clips using consistent prompts and references.
Specify aspect ratio (16:9, 9:16, 1:1) explicitly in the prompt for platform-optimized output.
Use clear, sequential instructions when editing videos to ensure the model interprets changes correctly.
Version history
Predecessor with shorter clips and no native audio.
Current — open Apache 2.0, 15s clips, native audio, four modes.
FAQ
Frequently asked questions
Wan 2.7 is Alibaba's open-weight video generation suite, released in April 2026. It includes text-to-video, image-to-video, reference-to-video, and video-to-video variants, all under Apache 2.0. It supports 1080p clips up to 15 seconds and is designed for production-ready, uncensored creative workflows.
On Venice, pricing starts at $0.55 for a 720p, 5-second clip. Prices scale with resolution and duration — a 1080p, 5-second clip costs $0.70. You pay per clip, with no subscription required.
Wan 2.7 is open-weight under the Apache 2.0 license, meaning the model weights are publicly available and can be self-hosted or modified. However, running it on Venice is a paid service based on usage.
No, Wan 2.7 does not support function calling or tool use. It is a pure video generation model focused on visual synthesis from text, image, or video inputs.
Wan 2.7 supports 1080p and 720p resolutions across all variants. Output is available in 16:9, 9:16, and 1:1 aspect ratios.
Only the reference-to-video variant (wan-2-7-reference-to-video) supports native audio generation. Text-to-video, image-to-video, and video-to-video variants do not include audio.
Yes. The wan-2-7-image-to-video variant animates still images into motion. It supports 720p and 1080p outputs with clip lengths of 5, 10, or 15 seconds.
Yes. The wan-2-7-reference-to-video variant allows you to maintain subject consistency using reference images. It also supports native audio and voice cloning, with clips up to 10 seconds long.
Wan 2.7 wins on flexibility, open licensing, and video editing. VEO 3.1 wins on cinematic quality and native audio generation with precise lip-sync. If you need commercial control and self-hosting, choose Wan 2.7. If you need hero-quality cinematic clips, VEO 3.1 may be better.
Run Wan 2.7 privately
No prompt logging. No data used for training.