We're excited to officially launch Hunyuan and Wan 2.1: the world's most powerful open source video generation models, right here on Civitai, with a motion-packed contest that'll keep you on the edge of your seat! (you should probably shift back, it's not good for your posture)
This is your chance to bring your cinematic genius to the table and prove you're not a robot by generating up some award-winning clips to make your audience feel [EMOTION]!
With text to video generations starting at just 100 buzz, there's never been a better time to dust off that director's chair, microwave up some popcorn, and get the camera GPU rolling! š¬šæš„šļøšļø
Hunyuan/WAN LoRA training coming soon.
Quick Model Rundown
Hunyuan [text to video]
Realistic cinematic content
Great for generating lifelike scenes, dramatic visuals, and environments with strong lighting and camera dynamics.High-quality image/video fidelity
Prioritizes visual sharpness and detail, making it ideal for short polished clips and concept teasers.Longer generation time, heavier hardware/buzz costs
Needs high VRAM and is slower per render, so better suited for intentional, high-effort outputs rather than rapid iteration.
Wan 2.1 [text to video, image to video]
Stylized or animated/cartoon-like content
Has flexibility for both realistic and stylized/cute/fun outputs, especially useful in meme, anime, or creator-focused videos.Fast iteration on limited hardware
Works on mid-high-range consumer GPUs, making it excellent for creators and hobbyists needing speed and accessibility.Creative experimentation
With support for image-to-video, video-to-audio, and editing, it's a versatile sandbox for creators doing remix, storytelling, or exploratory content.
š ļø How to Use Hunyuan / Wan 2.1
Check out theĀ Video Generation panel in theĀ Civitai Generator, and select Hunyuan (t2v) or Wan (t2v / i2v).
Check out ourĀ Education Hub guide how to best prompt with these models:
š Submission Guidelines
We're looking for videos that fully embrace the cinema theme, whether realistic, cartoon, or otherwise. Videos may leverage Hunyuan's t2v or Wan's t2v/i2v models and must be generated on Civitai (local generation provides too many potential advantages such as resolution, framerate, etc.). Hunyuan / Wan LoRAs can be used, so long as the result meets the contest criteria (PG/PG-13).
That said, this isn't a professional video editing/production contest (that's what Project Odyssey is for!): we're looking for the best examples of videos generated right from the source.
So pick a category below, draft up your prompts, and get generating!
Video Categories
š„ Action
Dynamic scenes that capture thrilling energy and excitement. Submissions could feature car chases, explosive sequences, martial arts combat, or daring stunts, depicted with high-octane movement, intense colors, and dramatic angles that pull viewers into the heart of the adventure.
š§ Horror
Visuals that evoke suspense, fear, or the supernatural. Abandoned buildings, haunted forests, eerie figures in shadow, or unsettling transformations depicted with stark contrast, desaturated color palettes, and compositions that suggest something lurking just out of frame.
Ā
š Romance
Intimate moments that celebrate connection and emotion. Sunset embraces, candlelit dinners, chance meetings in rain-soaked streets, or tender glances across crowded rooms, captured with soft lighting, warm color tones, and compositions that frame the subjects in ways that highlight their emotional bond.
š Comedy
Humorous scenes that spark laughter and delight. Submissions could feature slapstick mishaps, witty visual gags, absurd juxtapositions, or comedic character expressions, depicted with bright colors, exaggerated perspectives, and perfect timing that captures those priceless moments when hilarity ensues. Think colorful food fights, unexpected animal photobombs, ridiculous disguises, or the classic banana peel slip - all frozen at the exact moment that makes viewers grin!
š½ Sci-Fi
Futuristic visions that transport viewers beyond the ordinary. Advanced technology, alien worlds, space exploration, or dystopian societies depicted with sleek designs, otherworldly lighting effects, and striking contrasts between human elements and the extraordinary environments they inhabit.
š How to Enter
Simply head to theĀ video collection, click the blueĀ Submit an entryĀ button, select the applicable entry categoryĀ and video to submit!
ā Prizes
Everyone who participates will receive a film makers badge with elements of all these themes incorporated, and the winners will receive an exclusive animated directors cut version!
š„ Action
1st Place Entry - 100,000 ā”Buzz + Animated Cosmetic
2nd Place Entry - 50,000 ā”Buzz + Animated Cosmetic
3rd Place Entry - 25,000 ā”Buzz + Animated Cosmetic
š§ Horror
1st Place Entry - 100,000 ā”Buzz + Animated Cosmetic
2nd Place Entry - 50,000 ā”Buzz + Animated Cosmetic
3rd Place Entry - 25,000 ā”Buzz + Animated Cosmetic
ā¤ļø Romance
1st Place Entry - 100,000 ā”Buzz + Animated Cosmetic
2nd Place Entry - 50,000 ā”Buzz + Animated Cosmetic
3rd Place Entry - 25,000 ā”Buzz + Animated Cosmetic
š Comedy
1st Place Entry - 100,000 ā”Buzz + Animated Cosmetic
2nd Place Entry - 50,000 ā”Buzz + Animated Cosmetic
3rd Place Entry - 25,000 ā”Buzz + Animated Cosmetic
š½ Sci-Fi
1st Place Entry - 100,000 ā”Buzz + Animated Cosmetic
2nd Place Entry - 50,000 ā”Buzz + Animated Cosmetic
3rd Place Entry - 25,000 ā”Buzz + Animated Cosmetic
š Rules
All submitted videos must be created and submittedĀ after the contest start dateĀ ofĀ April 22nd, 2025, and before the end date ofĀ May 6, 2025, atĀ 23:59 CST. Previously created videos are not permitted.
Video Submission Limit: ParticipantsĀ may submit up toĀ 10 videosĀ for the contest.
Allowed Video Generation Services: VideosĀ must be created with theĀ Hunyuan or Wan 2.1 generation engine on Civitai.
"Starter" images for i2v submissionsĀ can be made with any model ecosystem (Flux, SD 3.5, SDXL, Illustrious, Pony, etc.) onsite or locally andĀ do not need to be uploaded to the collection, butĀ entries must be made with the Hunyuan or Wan 2.1 generation engines and have the appropriate tool tag (Hunyuan/Wan).
All submitted videosĀ must beĀ SFWĀ (PG/PG-13 ratings) and adhere to ourĀ Terms of Service.
All submissionsĀ must fit within one of the categories: Action, Horror, Romance, Comedy, Sci-Fi.
Winners will be selected via Staff-PickĀ after the closing date.
Each participant canĀ only win one categoryĀ in this contest. If you win in one category, you areĀ not eligibleĀ to win in another.
Contest manipulation and unfair practices, such as vote purchasing, usage of bot accounts, exploitation of technical vulnerabilities, or any other dishonest attempts to artificially influence results, are strictly prohibited. Participants found engaging in such activities will face immediate disqualification and a permanent ban from the website.
Generation Data Requirement:Ā All winners will be required toĀ post their generation dataĀ to claim their prize. You can submit videos without prompts and creation information for competitive reasons, but if you are selected as a winner, you must publish that information to claim your prize.
Level Up YOUR video game with Hunyuan/Wan!
Models / Prompts Used in Article
Lights Camera AI-ction Theater - Wan 2.1 i2v
Zooming in video capturing a dimly lit movie theater. The focal point is a large screen at the front of the theater, displaying the words "Lights, Camera, AI-ction!" in an elegant, cursive font that is illuminated in a bright, white color against a dark, navy blue background. The text is centered on the screen, creating a dramatic and cinematic effect. The theater seats are arranged in rows, facing the screen, with the nearest seats visible in the foreground. The seats are upholstered in a dark fabric, possibly red or brown, and are arranged in a classic, tiered layout. The aisle is visible, leading towards the screen, and the theater appears to be sparsely populated, suggesting that the audience is either waiting for the movie to start or has not yet arrived. The walls of the theater are adorned with warm, yellowish lighting, providing a cozy and inviting atmosphere. The lighting fixtures are mounted on the walls, adding to the ambiance. The overall setting suggests a traditional, possibly vintage-style movie theater, designed to evoke a sense of nostalgia and anticipation for the upcoming film.Action Samurai - Wan 2.1 t2v
Close up shot of the determined face of a battle-worn samurai. Camera pulls back to reveal him standing alone on a foggy battlefield, gripping his katana. Camera pulls back to reveal fallen warriors behind him. Wind whips through the trees, sending red autumn leaves swirling.Horror Monster - Wan 2.1 t2v
Close up shot of the blood-covered face of a frightening reptilian monster. Camera pulls back to reveal him standing alone in a foggy swamp, gripping his bloody claws.Romance Furry Couple - Hunyuan t2v
A happy furry anthropomorphic wolf couple having a picnic with wine, realistic styleComedy Jazz Dog - Wan 2.1 t2v
Video of A low angle shot of a jazz pianist in a dimly lit 1920s jazz bar, playing the piano with concentration. He wears a white shirt with suspenders and black trousers, his hands move rapidly on the keys. Camera pans left to low angle shot of a fluffy dog playing the trumpet.Sci-Fi Alien Sneaking - Wan 2.1 t2v
Dynamic shot of an alien sneaking around a derelict space station, camera panning around the alien. dramatic lighting, photorealistic detail.Level Up Robot - Wan 2.1 i2v
Video of a muscular, armored superhero in a dynamic, action pose raising his arms. The character is encased in a fiery, orange and yellow flame, with the word "LEVEL UP" prominently displayed in bold, white letters across the center. The background is dark and blurred, suggesting a dramatic, intense environment. The image has a video game interface aesthetic, with pixelated edges and a grid pattern, emphasizing a game-like progression.
Civit Logo - Wan 2.1 i2v
CGI video depicts the words "WAN21 ON CIVITAI" in large, illuminated, 3D letters standing on a rain-soaked, rocky ground. The letters are adorned with numerous small, glowing yellow bulbs, creating a festive, neon effect. The background is a foggy, dark forest with rain falling, adding a dramatic, moody atmosphere. The scene evokes a futuristic, dystopian vibe, with the glowing letters standing out starkly against the dark, wet environment. The overall color palette is dominated by dark greens, grays, and the warm glow of the bulbs.