Video Generation
Turn a prompt — or a still you already have — into a finished clip ready for Reels, TikTok, a product page, or a brand intro. Our lineup spans the flagships: Seedance 2.0 for cinematic text-to-video with native audio, Happy Horse 1.0 as a fast text-or-image starting point, Wan 2.7 I2V for animating stills, and Pruna for fast draft mode iterations. Pick the workflow that fits the brief and ship.
Seedance 2.0 Pro is ByteDance's flagship text-to-video model — strong prompt adherence, natural motion, and optional multimodal references (images, videos, audio) to lock identity, style, and rhythm. This is the native ByteDance ModelArk endpoint; the WaveSpeed-routed siblings stay in the catalogue for users that already point at them.
Seedance 2.5 Text to Video is ByteDance's next-generation video model, turning a written brief into a single continuous take of up to 30 seconds — with dialogue, effects and ambience generated in the same pass. It renders up to native 4K and, when you need it, follows up to 30 reference images plus video and audio references for characters, products and motion.
Wan 3.0 Prime Reference to Video is the premium tier of Alibaba's Wan 3.0 reference mode, aimed at final deliverables rather than drafts. It builds a video of up to 30 seconds, with sound, from your own images, clips and audio — and is the tier to reach for when a character, a product label or a brand mascot has to stay locked across a reference-heavy shot.
Wan 3.0 Reference to Video turns the material you already have — product shots, character photos, a sample clip, a voice or a jingle — into a new video of up to 30 seconds with sound generated in the same pass. It is the standard tier of Alibaba's unified Wan 3.0 model and takes up to ten images, five video clips and five audio clips per shot, at 480p, 720p or 1080p.
Wan 3.0 Video Edit rewrites an existing clip from a plain-language instruction — remove an object, change an outfit or a character's age, restyle or relight the whole shot, or even change what someone says. It is the editing mode of Alibaba's unified Wan 3.0 model, keeps the original framing and length, and can pull new looks from up to ten reference images.
Pruna Video Avatar is the fastest talking-avatar / lip-sync model on the market — drop in a portrait photo plus either an audio clip or a written script, and get back a video of the subject speaking with synchronised lip movement, in 30 voices and 10 languages.
Wan 2.7 Image-to-Video animates a still image (or continues an existing clip) into a 3-15 second video, with optional first-and-last-frame control and audio synchronisation. Output runs at 720p or 1080p.
Happy Horse 1.0 is Alibaba's text- and image-to-video model. Generate a clip from a prompt, or animate a single still image into a fluid 3–15 second video at 720p or 1080p, with five aspect ratios for vertical, landscape and square delivery.
Pruna Video is a fast, multi-modal video generator — text-to-video, image-to-video and audio-to-video from a single endpoint. A built-in **draft mode** cuts the cost to about a quarter for rapid iteration; flip it off when you're ready for the final delivery take.
Wan 2.2 i2v Fast is the PrunaAI-optimised variant of Alibaba's Wan 2.2 A14B image-to-video — engineered for short turnaround and predictable per-video pricing. Drop in a first frame, optionally a last frame, write a short motion prompt, and ship a 480p or 720p clip for ads, social cutdowns or product loops.
Wan 2.7 Video Edit is Alibaba's video-to-video editor — drop in a 2-10 second clip plus a written instruction ("swap the weather", "restyle as claymation", "replace the backdrop") and the model rewrites the clip while preserving the camera move, framing and subject motion of the original.
Generate video that actually moves the brand
Text-to-video, image-to-video, with native audio when you want it. One prompt, four flagships.
Frequently asked questions
Text-to-video or image-to-video — which workflow should I use?
Use text-to-video (Seedance 2.0, Happy Horse 1.0, Pruna Video) when you're starting from a written brief and want the model to compose the whole scene. Use image-to-video (Wan 2.7 I2V) when you already have the exact frame — a product photo, a hero shot, a character design — and want it animated faithfully.
Which model should I start with?
For commercial work — product demos, ads, brand intros — start with Seedance 2.0 for the cleanest cinematic motion and native audio in one pass. For fast iteration on social-format clips, Happy Horse 1.0 is the quickest path. For animating a specific still you've already approved, Wan 2.7 I2V.
How long can a generated clip be?
Most models in this category top out around 5–10 seconds per generation. For longer pieces, generate in shots and stitch them — using a consistent opening frame as the i2v anchor keeps subject and lighting stable across cuts.
Does any model generate audio along with the video?
Seedance 2.0 supports native audio generation in the same pass — ambient sound, foley, and basic music matched to the scene. The other models output silent video; layer audio in post if you need it.
Best for short-form social — Reels, TikTok, Shorts?
Generate at 9:16 aspect ratio and keep duration between 5 and 8 seconds for tight loops. Happy Horse 1.0 and Pruna Video are the fastest options for the high iteration counts these formats demand; Seedance 2.0 wins on the final hero post when you want polish.
Can I use the generated videos commercially?
Yes. Every model in this category supports commercial reuse under our terms, including ads, product pages, social campaigns and branded content. Outputs are yours to publish — no per-clip royalty, no attribution requirement.
Compare in Studio
Open Studio with video generation preselected and A/B Seedance, Happy Horse and Wan 2.7 on the same prompt.
Launch Studio →Try Seedance 2.0
The flagship for cinematic text-to-video with native audio — the one we reach for on hero posts.
Open the model →