What is Wan 2.6 (V2V)?

Wan 2.6 (V2V) is an AI model (video to video) available on Artifio.ai through Wan. You can run it in our workspace from a single shared wallet and download your outputs.

How much does Wan 2.6 (V2V) cost on Artifio?

Wan 2.6 (V2V) starts at about $0.350 per generation on Artifio.ai. The exact cost is shown in the workspace before you confirm and can vary with the options you select.

How do I use Wan 2.6 (V2V) on Artifio?

Click Generate Now to open the workspace at /create/wan-2-6-video-to-video. Fill in the form fields, review the live cost estimate, and run your generation.

What kind of outputs does Wan 2.6 (V2V) produce?

Wan 2.6 (V2V) is configured for video to video workflows in the video category. See the technical parameters below for supported inputs and options.

Can I try Wan 2.6 (V2V) for free?

Sign up free, top up your wallet from $5, and you can run Wan 2.6 (V2V) pay-as-you-go without a separate subscription for this model.

Wan 2.6 (V2V) — AI Video Generator

Name: Wan 2.6 (V2V)
Brand: Artifio
Price: 0.350 USD
Availability: InStock

$1 on us$1 on us + a free Gold trial to get you startedClaim your $1 + trial Terms apply

artifio.ai

Wan 2.6 (V2V) — AI Video Generator | Artifio.ai

About Wan 2.6 (V2V)

Wan 2.6 (V2V) is the video-driven mode of Alibaba's Wan 2.6 series, built on the reference-to-video model (Wan2.6-R2V) introduced with the series on December 16, 2025. You upload a reference video that carries a character's appearance and voice, then write a text prompt, and the model generates new scenes starring that same character. Alibaba describes it as China's first reference-to-video generation model and framed the launch around letting anyone appear in AI-generated videos as themselves, in their own voice.

Like the rest of the 2.6 generation it is a hosted model, available through Alibaba Cloud's Model Studio and the Wan website rather than as open weights. Artifio runs it beside 100+ other models in one workspace with per-generation billing.

Features

Takes a reference video as input and extracts both the subject's appearance and voice for reuse in new footage.
Generates new scenes from a text prompt with the referenced character preserved across shots.
Outputs 720p or 1080p at 30 fps, with audio generated alongside the picture.
Clip lengths run from 2 to 10 seconds in one-second steps.
Benefits from the 2.6 series' multi-shot storytelling and improved audio-visual synchronization.

What people use it for

Casting yourself or a presenter into scenes that were never filmed.
Keeping a recurring character consistent across an episodic content series.
Producing spokesperson-style ads from a single reference clip of the talent.
Generating dialogue scenes where the referenced person's voice must carry over.

Reported strengths

Identity and voice continuity come from one reference clip, with no fine-tuning step.
Audio is part of the same generation pass, so the output speaks in the referenced voice immediately.
It was the first model of its kind from a Chinese lab by Alibaba's own account, and the mode matured further in Wan 2.7.

Reported limitations

The duration ceiling is 10 seconds, shorter than the 15 seconds offered by the series' text-to-video and image-to-video modes.
Resolution is capped at 1080p.
No open weights exist for Wan 2.6; the Wan GitHub organization's newest open release is still Wan 2.2.

Sources

Alibaba unveils Wan2.6 series enabling everyone to star in videos (Alibaba Cloud)
Video generation, Alibaba Cloud Model Studio documentation (Alibaba Cloud)
Wan2.2: open and advanced large-scale video generative models (GitHub (Wan-Video))