Alibaba Cloud tools

HappyHorse 1.1 is a video generation model from Alibaba.

Wan 2.7 is a video generation model from Alibaba.

Wan 3.0 Prime is a fast-mode variant of [Wan 3.0](https://openrouter.ai/alibaba/wan-3.0) from Alibaba. It supports text-to-video and first-frame image-to-video generation.

Generate 480p, 720p, or 1080p video from a text prompt with the Wan 3.0 model.

Generate 480p, 720p, or 1080p video from a first-frame image with the Wan 3.0 model.

HappyHorse 1.0 is a video generation model from Alibaba.

Alibaba's most advanced video generation model, supporting over 10 visual creation capabilities in a unified system.

Generate a new scripted scene that preserves character appearance and voice from reference images and video clips (up to 5 combined) — multi-entity: each reference can carry its own voice sample, and the prompt addresses them by ordinal ("Image 1 walks past Video 1..."). Supports an optional pinned first frame, a nine-panel storyboard image as the sole reference, negative prompts, and prompt rewriting. Returns a downloadable MP4 video_url (24h expiry), 2-15 seconds (2-10 with a reference video), 720P/1080P. Suited for: consistent-character series, dialogue scenes, virtual presenters, storyboard-to-video.

Generate images from text, edit or compose from up to nine input images, place objects with bounding boxes (interactive editing), or produce a story-coherent SET of up to 12 images from one request — the full-featured tier — up to 4K text-to-image, interactive bounding-box editing, custom color palettes. Supports thinking mode, custom color palettes (3-10 colors), and free width*height sizing. Returns PNG image URLs (24h expiry). Suited for: seasonal/campaign image series with a consistent subject, product placement into scenes, style-guided redesigns, high-resolution art and print assets.

Generate images from a text prompt, or edit/compose from up to three input images with a natural-language instruction (one model does both) — the flagship tier — dense information layouts, text as small as 10px, native rendering of 12 languages and 20+ fonts. Supports precise in-image text rendering (multi-line layouts, paragraphs, UI mockups, posters, slides), negative prompts, free width*height sizing up to 2048*2048 at 1:8-8:1, and 1-6 outputs per run. Returns PNG image URLs (24h expiry). Suited for: marketing posters and banners, UI/slide mockups, storyboards, product shots, style transfer and subject edits.

Generate images from a text prompt, or edit/compose from up to three input images with a natural-language instruction (one model does both) — the standard tier — balances quality and speed. Supports precise in-image text rendering (multi-line layouts, paragraphs, UI mockups, posters, slides), negative prompts, free width*height sizing up to 2048*2048 at 1:8-8:1, and 1-6 outputs per run. Returns PNG image URLs (24h expiry). Suited for: marketing posters and banners, UI/slide mockups, storyboards, product shots, style transfer and subject edits.

Generate images from text, edit or compose from up to nine input images, place objects with bounding boxes (interactive editing), or produce a story-coherent SET of up to 12 images from one request — the faster tier — same feature set up to 2K. Supports thinking mode, custom color palettes (3-10 colors), and free width*height sizing. Returns PNG image URLs (24h expiry). Suited for: seasonal/campaign image series with a consistent subject, product placement into scenes, style-guided redesigns, high-resolution art and print assets.

Returns companyId, countryCode, currency, image, images, isSponsored, moq, price, priceText, productId, rating, reviewCount, supplierName, supplierUrl, supplierYears, title, url.