Search the full model catalog.

Renders a text prompt with the FLUX.1 checkpoint you pick, such as FLUX.1 [dev], [schnell] or Krea [dev], or with one of your private FLUX.1 fine-tunes.

FLUX.1 Dev is Black Forest Labs' 12B open-weight text-to-image model, distilled from FLUX.1 [pro]. It can spell short text inside the image, like the words on a sign or a label.

Removes the background from photos and returns each subject as a PNG cutout with transparency. One run takes up to 30 images.

Sets a still image in motion, following a text prompt for what should move. The model is I2VGen-XL from Alibaba's Tongyi Lab, with adjustable frame rate and output size.

Make your portrait alive! It is a tool that turns your portrait into a video using a provided reference video.

It generates a video from the given prompt.

Places the subject of one photo onto a background image you choose and relights the subject to match the new scene, using IC-Light's background-conditioned model.

More Lighting! IC-Light is used to manipulate the illumination of images using a text-conditioned model.

An image generator that takes its layout from you: captioned boxes drawn on a canvas plus a background prompt. LMD+, from the LLM-grounded Diffusion paper, keeps each object inside its box.

hair-fast-gan applies the hairstyle and hair color to the input image.

Restores blurry, low-resolution photos: CodeFormer rebuilds the face nearest the center and Real-ESRGAN enlarges the whole frame up to 10 times.

It creates a video by combining the input image with the movements of a person in a video and adding the audio from the input audio.

It creates a video by incorporating the movements and audio of a person from a given video into the input image.

Transform your 2D object into 3D.

Reads text aloud in the voice of a short reference clip. Built on Coqui's XTTS-v2, it needs about six seconds of speech, and the text can be in any of 17 languages, whatever language the clip is in.

OpenAI Whisper Medium transcribes speech to text with timestamps and optional speaker diarization. Works across many languages and noisy audio.

Unlimited OCR by Baidu parses scanned pages and long PDFs in one pass and returns structured, readable text for search, RAG, and document workflows.

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.

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.

A 3B OCR model from rednote-hilab that turns PDFs and images into Markdown with HTML tables and LaTeX formulas, plus a labeled box for every block. Its makers designed it to read almost any human script.