Wan 2.6 is the latest generative video model from Alibaba DashScope — and it powers HappyHorse. Run it in your browser for text-to-video and image-to-video, free.
HappyHorse runs on Wan 2.6 from Alibaba DashScope. You get the latest model without setting up Cloud accounts.
Describe any scene in a sentence and Wan 2.6 generates a cinematic clip — realistic motion, lighting, and camera moves.
Upload a still photo and Wan 2.6 animates it with natural motion — perfect for product shots and portraits.
Export Wan 2.6 clips at 720p or 1080p. Clips run from 5 to 15 seconds in 16:9, 9:16, or 1:1.
Wan 2.6 follows complex prompts better than earlier versions — camera direction, subject count, and scene details all land more reliably.
New HappyHorse accounts receive free trial credits for Wan 2.6. No credit card required.
Three steps from idea to Wan 2.6 video.
Open HappyHorse and choose text-to-video or image-to-video. Wan 2.6 is selected by default.
Type your scene description or upload a reference image. Choose 720p or 1080p, aspect ratio, and duration (5 to 15s).
Hit Generate. Wan 2.6 renders in the cloud and the finished clip lands in your library ready to download.
Wan 2.6 is the current flagship of Alibaba DashScope's Wan video model family. Compared to Wan 2.5 and Wan 2.1, version 2.6 brings better motion realism, sharper detail, improved temporal consistency, and noticeably stronger prompt adherence — especially for multi-subject scenes and camera direction. HappyHorse exposes Wan 2.6 through a simple editor that runs in your browser. No API keys, no DashScope billing, no Python — just prompt, click Generate, and get a clip.
• Modes: text-to-video, image-to-video, video-to-video • Resolution: 720p and 1080p • Duration: 5, 10, or 15 seconds • Aspect ratio: 16:9, 9:16, 1:1 • Input languages: English, Chinese, and most major languages If you want prompt-only generation, use /text-to-video. For animating a still, use /ai-video-from-image. Both routes use Wan 2.6 under the hood.
Wan 2.6 rewards specific, visual prompts. A strong structure: • Subject: "a red vintage sports car" • Action: "drifting through a wet neon alley" • Camera: "tracking shot, low angle" • Style: "cinematic, film grain, 35mm" Keep prompts under 60 words, avoid contradictory instructions, and describe what you want to see rather than what you do not want. For image-to-video, the prompt should describe the motion you want to add to the existing image, not the image itself.
Yes — new accounts get free trial credits to run Wan 2.6. For higher volume, paid plans are available. See /pricing for details.
No. HappyHorse runs entirely in the browser. No downloads, no API keys, no DashScope setup.
Up to 15 seconds per clip on HappyHorse, at 720p or 1080p.
Wan 2.6 has improved motion quality, better prompt adherence, and stronger temporal consistency compared to 2.5.
Yes. Upload a still image and Wan 2.6 animates it. Use /ai-video-from-image or the image-to-video mode in the editor.
Clips generated on a paid HappyHorse plan can be used in commercial content including ads and client work.
More ways to create, edit, and enhance video with HappyHorse.
Generate videos with the Wan model family.
Try itNano Banana model for fast iteration.
Try itTurn text and images into HD video in seconds.
Try itGenerate AI videos from a simple text prompt.
Try itAnimate any still image into a dynamic clip.
Try itUpload a photo, get a motion video back.
Try itUpscale videos to HD/4K with one click.
Try itExtend short clips with AI-generated continuation.
Try itWan 2.6 is the latest generative video model from Alibaba DashScope — and it powers HappyHorse. Run it in your browser for text-to-video and image-to-video, free.