Search the full model catalog.

Dresses the person in one photo in up to two garments or accessories from other photos. Keep the original scene or restage it in a studio, indoor or outdoor setting, and get a still image, a short clip or both.

Google's Veo 3.1 Fast turns a text prompt, optionally with a first frame or a first and last frame, into a 4 to 8 second video with dialogue, sound effects and ambient audio, at 720p, 1080p or 4K.

Access the Veo 3.1 API for high quality video generation with audio. View pricing, integration docs, and demo videos to start creating advanced scenes with simple API calls.

Moondream 3 Preview Detect takes an image and a short text description and returns a bounding box for each matching object, as normalized x_min, y_min, x_max and y_max values in JSON.

Name an object and Moondream 3 Preview returns one normalized x, y point for each match, as JSON you can count or draw as markers on the image.

Moondream 3 Preview writes a prose caption for an image. There is no prompt: you choose Short, Normal or Long and the model decides what to describe.

Ask Moondream 3 Preview a question about one image and get the answer as text or JSON. A reasoning step runs by default and can be switched off for simple questions.

Ovi generates a short video and its soundtrack together from a text prompt and an optional first-frame image, with spoken lines you write between speech tags.

Lucy Edit Dev applies a written instruction to an existing video, such as swapping an outfit or turning a person into a creature, while the original motion and framing stay in place. Common edits need no masks.

Puts people and pets from one to three photos into one of 30 preset scenes and animates the result, such as a couple in a Polaroid frame or a person posing with a pet.

Gemini 2.5 TTS runs Google's Gemini 2.5 Flash Preview TTS model: write a line with a plain-language direction such as "say cheerfully" and get it spoken in one of 30 prebuilt voices.

Hunyuan Flux SRPO is FLUX.1 dev (12B) fine-tuned by Tencent Hunyuan with SRPO, a human-preference training method aimed at realism and aesthetic quality without oversaturated color.

Puts two people from separate photos into one Polaroid-style flash snapshot and animates them with a chosen gesture, such as a hug or a high five. The animation runs on Wan 2.2.

ByteDance image model that generates and edits pictures from a prompt. It takes up to 14 reference images and can return a set of up to 15 related images in one run.

Generates a 360-degree equirectangular panorama from a text prompt, blending the left and right edges so they join when the image wraps around. Tencent built it on FLUX.1 [dev].

Builds a full 360-degree equirectangular panorama around a single photo, generating the parts of the scene the camera never captured. You set the field of view the photo covers on the sphere.

Turns one portrait photo into a short Wan 2.2 clip with one of 20 preset effects, ranging from costume and superhero transformations to reactions like crying or laughing.

Restages one photo in a themed scene chosen from more than 250 named effects and animates the result into a short clip. You can get the still, the video or both.

Edits a photo from a plain-language instruction and can take up to two more images as references, for example a logo to print on a shirt. With no image attached, it creates one from the prompt.

OpenAI's smaller open-weight reasoning model, a mixture-of-experts with 21B parameters (3.6B active) under Apache 2.0, served here as a text chat model with per-session history.