# Wan 2.6 (V2V)

> Wan 2.6 (V2V) is available on Artifio.ai — a unified AI workspace with a single pay-as-you-go wallet.

- Canonical page: https://artifio.ai/models/wan-2-6-video-to-video
- Run it: https://artifio.ai/create/wan-2-6-video-to-video
- Modality: video to video
- Model brand: Wan
- Pricing: from about $0.350 per generation (exact cost shown in the workspace before you confirm; varies with selected options)
- Typical generation time: ~800s (catalogue estimate)
- Full pricing table: https://artifio.ai/pricing/models

## 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

1. Alibaba unveils Wan2.6 series enabling everyone to star in videos (Alibaba Cloud): https://www.alibabacloud.com/blog/alibaba-unveils-wan2-6-series-enabling-everyone-to-star-in-videos_602742
2. Video generation, Alibaba Cloud Model Studio documentation (Alibaba Cloud): https://www.alibabacloud.com/help/en/model-studio/use-video-generation
3. Wan2.2: open and advanced large-scale video generative models (GitHub (Wan-Video)): https://github.com/Wan-Video/Wan2.2

---

About Artifio.ai: 100+ AI image, video, and audio models in one workspace with a single shared wallet — no per-model subscriptions. This is the markdown rendition of https://artifio.ai/models/wan-2-6-video-to-video for AI assistants and crawlers; the HTML page is canonical.