Everything in the catalog

Shows how a garment looks on a model with AI virtual try-on.

Generates images from text prompts with the OpenDalle V1.1 model.

Reads your text aloud as an audio file, with an optional short clip to match the voice style.

Generates FLUX images in only 8 steps with the Hyper-SD LoRA.

An image generation hub that runs Flux, SD 3.5 and LoRA models without needing a local GPU.

Estimates a depth map from a single photo.

Ask questions about images with the DeepSeek-VL2 small vision-language model.

Generates MIDI music with a transformer, optionally continuing a MIDI you provide.

Anime-character style Japanese text-to-speech with many speaker voices.

Converts a voice recording to another voice using a VITS-based voice conversion framework.

Estimates how much GPU memory a language model needs for a given size, quantisation and context length.

Search a library of 3D assets by meaning and preview them with simulation renders in USD format.

Four-times image upscaling with the AuraSR GAN super-resolution model.

Converts an ebook into an audiobook, with support for more than 1,100 languages.

Removes backgrounds from anime and illustration images using a model trained on that style.

Try a set of Qwen3-VL models on image and video questions in one place.

Edit images with Qwen, then enhance the result with FLUX.2 Klein, in one image editing app.

Describes images with the Florence-2 large and base models.

Text-to-speech app that clones a voice for the MegaTTS 3 model, so generated speech follows a chosen voice.

Streams live microphone audio through Whisper on the GPU in your browser and prints the words as you speak.

English speech recognition with NVIDIA Parakeet, returning text and timestamps.

Clone a voice and speak text with it in many languages, including English, Chinese, Japanese, German and Spanish.

Shows captions from several image captioning models side by side for the same picture.

Swaps the background of a product photo for a generated scene described by a prompt.