Everything in the catalog

Transcribes speech as you talk using Voxtral Mini, entirely in your browser.

Few-step text-to-image generation with SDXL-Lightning.

Converts PyTorch model weights to the safetensors format; asks for a Hugging Face token.

Replaces a character in a video clip with a different character taken from a reference image, using a MiniMax-H3 LoRA.

Convert page images into structured document markup with a small vision-language model.

Text-to-speech demo of Fish Audio Drama 3 that streams generated speech, with voice cloning among its listed features.

Speech recognition with the smaller Distil-Whisper model, executed locally in the browser.

Generate and complete code by typing a prompt into the Code Llama playground.

Generates and edits images from text in eight steps using the Qwen-Image-2.1-Turbo model.

Text-to-image with Tencent's HunyuanDiT, which accepts Chinese and English prompts.

Text-to-speech covering a very large range of languages.

Fast 4B-parameter FLUX.2 model for text-to-image and editing.

Browser-based depth estimation that turns a photo into a depth map using Transformers.js.

Text-to-speech demo of the MetaVoice 1B model, which turns typed text into spoken audio.

Describe or question images and videos using Qwen2-VL 7B.

Pulls text out of images and documents, letting you pick among several OCR models such as SmolDocling and Nanonets.

Experimental image editor built on FLUX.2 Klein 9B with a choice of LoRA adapters.

Creates a comic from a single text prompt.

Extractive summarizer that picks key words, sentences or paragraphs from a text.

Turns a written script with speaker tags into spoken dialogue audio using the Dia 1.6B model.

Background remover with a choice of rembg models; outputs the cutout image and the mask separately.

Text-to-speech from Resemble AI's Chatterbox model, able to mimic a voice from a reference clip.

Paint over an unwanted object or blemish in a photo and the LaMa model fills in the area.

Edit and manipulate images with AI using Kontext adapters.