Search the full model catalog.

Short film moments built from a portrait: a rain-streaked window, a homecoming hug, a forest wedding, a minidrama reveal. Thirteen of the 35 scenes take a second photo, of another person or a pet.

A superpower effect for portraits: the subject commands fire, ice, lightning, magma, wind, metal or time in a short VFX shot. All 13 scenes work from a single photo.

Turns one portrait into a short staged performance clip. The 25 preset scenes run from a K-hip hop rooftop set and a Hanfu dance hall to a flamenco dancer with a guitarist on a slackline between cliffs.

Turns a portrait or a pet photo into a short comedy clip with a twist. The 18 scenarios cover deadpan gags, reveal jokes and pet stories, and a few of them cast two people.

Puts the person in a portrait into 12 photoreal antigravity scenes, from a temple meditation to a damaged orbital station. In most they float; in a few, pool water, a TV or a skyline moves around them.

Restyles the person in a portrait as an anime, manga, ink-painting or pixel-art character in a short video with sound. About half of the 48 scenarios start in live action and switch to a drawn style.

Casts the person in one portrait photo as the lead of a family-film style comedy, often beside an animated co-star like a pigeon or a cat. 17 storylines, up to 15 seconds with sound, at 720p or 1080p.

Qwen3.6-27B is a 27B dense vision-language model from Qwen for agentic coding and reasoning, with optional thinking traces in outputs. On Wiro a request takes about 25K input tokens and returns up to 16K.

Multilingual text-to-speech from OpenBMB: 2B parameters, 30 languages, 48 kHz output. Clone a voice from a short clip or describe a new one in words, with audio streamed back as it is generated.

Generate or edit images with GPT Image 2 from OpenAI. It delivers strong instruction following, sharp text rendering, and flexible sizing up to 4K.

Seedance 2.0 Fast R2V is the lighter variant of ByteDance's Seedance 2.0 for reference-to-video, with the same video, image and audio references and editing tasks, rendered at 480p or 720p in 4 to 15 second clips.

Snap your fingers and the world around you freezes in a cinematic time-stop video made from one portrait. 63 scenarios across street magic, bullet-time action, fantasy, supernatural, sports and weddings.

Generate 2-10s videos in 720p or 1080p from reference images or a short clip. Wan 2.7 R2V keeps subject identity while following your prompt.

Seedance 2.0 Fast by ByteDance generates short MP4 videos faster and cheaper than Seedance 2.0, with the same multimodal capabilities, limited to 480p / 720p output.

Reconstruct a 3D scene from multi-view photos or a short video. Produces 3D Gaussian splats with depth maps and cameras, plus a colored point cloud.

Seedance 2.0 R2V builds a new clip with synced audio from 1 to 3 reference videos, reusing their motion and camera work, with optional reference images and audio and output up to 4K and 15 seconds.

Seedance 2.0 by ByteDance generates short MP4 videos from text, with optional image and audio references. It supports synced audio, stable motion, and cinematic camera direction.

The distilled 8-step checkpoint of Baidu's ERNIE-Image text-to-image model. Baidu says it keeps the parent's text rendering and layout handling, and it includes a prompt enhancer.

Multilingual speech-to-text for 25 European languages with auto language detection, punctuation, capitalization, and optional timestamps.

Baidu's 8B open-weight text-to-image model, the 50-step SFT release of ERNIE-Image. Its strength is images that carry a lot of text, such as posters, infographics and multi-panel comics.