Magi-1.video

Wan 2.2

Click or drop image here
Support PNG, JPG, JPEG (max 10MB)

Experience the Next Frontier of AI Video with Wan 2.2

The Wan 2.2 model is engineered to provide creators with an unprecedented level of control, aesthetic quality, and motion fidelity. Wan 2.2 empowers you to produce complex, emotionally resonant, and visually stunning videos that precisely match your creative intent. Forget the limitations of previous models; Wan 2.2 understands nuance, masters complexity, and delivers cinematic excellence.

MoE Architecture for Scalable Capacity

Wan 2.2 introduces a Mixture‑of‑Experts architecture across timesteps—specialized denoising experts handle different phases of generation. This increases effective model capacity while keeping computation cost stable, enabling larger total parameter scale without more runtime load.

Cinematic‑Level Aesthetic Control

Built with richly labeled aesthetic training data—lighting, composition, color tone, contrast—Wan 2.2 delivers controllable cinematic styles. Fine‑tune prompts to adjust mood, palette, framing and achieve video that matches your visual narrative intent.

Fluid Complex Motion Generation

Trained with over +83 % more video data than earlier versions, Wan 2.2 excels at recreating sweeping camera moves, object rotations, multi‑object interactions and choreography—preserving smooth arcs, realistic physics and seamless transitions.

How to Use Wan 2.2

Get started in just a few clear steps—no technical experience required.

1

Input Your Prompt

Write a text description specifying scene elements, lighting style, motion direction, composition, color tone and any desired objects.

2

Add a Source Image (Optional)

For image-to-video generation, upload a starting image to guide the composition and style of your creation.

3

Generate & Refine

Click generate and watch Wan 2.2 craft your video. Review the output and easily tweak your prompt to refine the result until it's perfect.

Explore the Power of Wan 2.2

Seeing is believing. Browse our gallery of examples to witness the stunning quality, fluid motion, and cinematic depth achievable with the Wan 2.2 model.

Close-up shot with soft lighting, focusing sharply on the lower half of a young woman's face. Her lips are slightly parted as she blows an enormous bubblegum bubble. The bubble is semi-transparent, shimmering gently under the light, and surprisingly contains a miniature aquarium inside, where two orange-and-white goldfish slowly swim, their fins delicately fluttering as if in an aquatic universe. The background is a pure light blue color.

Under a vast azure sky, illuminated by the gentle and warm sunlight from the side, a red-haired woman was smiling and laughing joyfully. Her long, curly tresses dance in the breeze. Dressed in a green suit adorned with floral patterns and fitted trousers, she pairs her outfit with striking neon green ankle boots. A large-brimmed straw hat, slightly drooping at the edges, crowns her head. Standing on a rural path blanketed in golden hay, expansive fields and a pristine blue horizon form the backdrop. With hands aloft, she wields a blue garden hose from which a cascade of multicolored flowers erupts instead of water, scattering like fireworks in the air. The blossoms, diverse in hue and shape, gleam with a gentle luster under the sun's rays.

A surreal digital artwork features a circular grass tunnel extending infinitely, centered on a deep blue sky. Three golden-haired figures in black coats, light skirts, and boots walk or fall along the tunnel’s inner walls, defying gravity in a cyclical motion. Distorted perspective and repetition create a dreamlike, endless loop, evoking a rabbit-hole fall, blending reality and imagination with avant-garde flair.

In an underwater ballroom, transparent jellyfish glow gently between marble columns. Couples in formal attire waltz gracefully in the center of the dance floor—men in black tuxedos, women in flowing long dresses, their outfits swaying softly in the gentle currents. Sea turtles swim slowly overhead, their reflections cast on the smooth coral floor. The camera moves slowly to the right, capturing this dreamlike scene.

Documentary handheld camera style, three missiles ascend vertically from the ground, emitting large amounts of flame and smoke from their tails, showcasing powerful thrust. The background is the Gobi Desert engulfed in a massive yellow sandstorm, with the outline of distant mountains barely visible. Some vehicles and equipment used for missile launches are visible on the ground. In the night scene, a shallow depth of field effect makes the subject stand out more, with a striking contrast between warm and cool tones. In a close-up shot, the missiles slowly rise; a tilted composition captures the awe-inspiring moment of the missile launch.

In a surreal video, four miniature skiers glide down a winding, three-dimensional trail of thick white paint on a plain white canvas-like background. The textured paint mimics snow, with visible brushstrokes and uneven edges, enhanced by light and shadow. The skiers, in colorful gear, are posed dynamically from top to bottom, each casting a shadow that heightens the illusion of depth. This scene miniaturizes a grand outdoor sport into a vivid, imaginative artwork.

Frequently Asked Questions about Wan 2.2

Have questions? We have answers. Find out everything you need to know about the Wan 2.2 video model.

Wan 2.2 is a state-of-the-art, foundational text-to-video and image-to-video AI model. It is designed to generate high-quality, cinematically styled videos with a strong focus on precise prompt following, complex motion, and aesthetic control.

Wan 2.2 introduces several key innovations: a Mixture-of-Experts (MoE) architecture for better prompt understanding, training on a significantly larger and more diverse dataset for improved motion and generalization, and a focus on labeled aesthetic data for cinematic control. This makes the Wan 2.2 model far more capable and precise.

Yes. Wan 2.2 was trained on aesthetic‑labeled data. You can adjust lighting, contrast, color tone and framing via prompts or UI controls to guide visual style.

The open-source Wan 2.2 model is optimized to generate video at 720P resolution (1280x720 pixels) and a smooth 24 frames per second (fps), which is the standard for cinematic film.

With over +83 % more video training data, Wan 2.2 generates fluid object and camera motion—including rotations, choreography and scene transitions—while preserving realism.

Wan 2.2 supports both text‑to‑video (T2V) and image‑to‑video (I2V) workflows—allowing custom visuals from text prompts or reference images.

Yes. Its cinematic narrative control, precise motion and aesthetic tuning are ideal for architectural walkthroughs, medical visualization, fashion motion clips, sports analysis and more.

Yes. Wan 2.2 is released as open‑source foundation models under Apache‑2.0. You can integrate it into research or industrial pipelines with full code and weight access.

Ready to Create Cinematic Videos?

Experience Wan 2.2’s unmatched combination of motion realism, stylistic control, and GPU efficiency. Start building professional‑grade videos today.