Make, edit and restore images.

Text to image, inpainting, instruction edits, photo restore and speech. Pick a tool to get started.

Generate

Generate images from a text prompt using SD 1.5 or SDXL models.

Generate images from a text prompt using FLUX.1-dev (Nunchaku).

Generate images from a text prompt using Qwen Image (Nunchaku, lightning 4/8-step).

Edit an existing image with a text prompt using Flux.1 Kontext (Nunchaku).

Instruction-based image editing powered by Qwen Image Edit (Nunchaku).

Paint over part of an image and describe what should go there (SD 1.5 / SDXL).

Extend an image past its edges — zoom out or widen the frame (SD 1.5 / SDXL inpaint).

Colorize black & white photos, upscale 2×/4× and restore faces.

Change the lighting of a photo: describe the light and pick where it comes from (IC-Light).

Remove backgrounds, turn an image into a prompt, or make a 3D parallax photo from depth.

Turn photos of an object (front, plus optional sides and back) into a 3D model (Hunyuan3D-2mv).

Turn text into speech with Kokoro voices.

Split a song into vocals, drums, bass and the rest, or take the background noise out of speech.

Run several tools in a row with one click — e.g. generate → cut out as a sticker, or photo → vector logo. Save your own.