r-k-jo

You can also target WASM. Depending on the model size, it can be really good. Here are some examples of quantized models.

Vision Model https://huggingface.co/spaces/radames/Candle-Moondream-2 Blip Image Captioning https://huggingface.co/spaces/radames/Candle-BLIP-Image-Capt... Microsoft Phi 2 https://huggingface.co/spaces/radames/Candle-phi1-phi2-wasm-...

Clipper.js is built on top of Mozilla's Readability library, Turndown to convert HTML to Markdown https://github.com/philschmid/clipper.js

It's using ~15GB VRAM

I also made a demo with Gradio, but it's 2x slower than fal.ai! Using stable-fast compilation running on a single A10G

https://huggingface.co/spaces/radames/Real-Time-Text-to-Imag...

I you have GPU/cuda/Docker you can try it locally

docker run -it -p 7860:7860 --platform=linux/amd64 --gpus all -e SFAST_COMPILE="1" -e USE_TAESD="0" registry.hf.space/radames-real-time-text-to-image-sdxl-lightning:latest python app.py

Here is a collection of demos with fast LCM on HuggingFace

https://huggingface.co/collections/latent-consistency/latent...

yes, just publish it on the readme.me here https://github.com/radames/Real-Time-Latent-Consistency-Mode...

Hi, I've built the demo. Unfortunately, it's running on a single GPU and can only be used concurrently by a few users. For a better real-time experience, you would need a dedicated machine.

You can get the source here and how to run: https://github.com/radames/Real-Time-Latent-Consistency-Mode... See video here: https://twitter.com/radamar/status/1718783886413709542 or on the Github readme.

FYI, this is made possible due to a new technique: https://latent-consistency-models.github.io, fine-tuning an existing models. The author will soon publish the training script. We'll see all the cool image models running at this speed! I'm excited for this! I can see many interesting experiments and projects emerging from this.

Powered by Candle Rust compiled to Wasm: https://github.com/huggingface/candle

Here's a live demo on HuggingFace https://huggingface.co/spaces/SpacesExamples/Fooocus

Yes, that's an unofficial implementation. The official implementation seems faster. Here is a demo for it https://huggingface.co/spaces/radames/DragGan"

Here is a live demo https://huggingface.co/spaces/radames/DragGan

Hugging face hub Python library has a Rust package to accelerate downloads https://github.com/huggingface/hf_transfer

> This library is a power user tool, to go beyond ~500MB/s on very high bandwidth network, where Python cannot cap out the available bandwidth.

Amazing project! Here is a demo including an input for negative prompt. It's impressive how it works. You can try:

prompt: relaxing jazz melody bass music negative_prompt: piano music

https://huggingface.co/spaces/radames/spectrogram-to-music

hi thanks for building this! Could you also enable a simple search by matching exactly words from the caption text? rather than semantic similarity?

Hi Sam, I'm big fan of your work! Coincidently, I just made a simple POC video editor by editing text using this speech to text model https://huggingface.co/facebook/wav2vec2-large-960h-lv60-sel.... It might be cool to integrate into your Videogrep tool, it also works offline with CPU or GPU, and gives you timestamps for word or character level.

https://twitter.com/radamar/status/1528660661097467904

No worries, it seems like using tensorflow.js and running locally on your browser.

https://github.com/tensorflow/tfjs-models/tree/master/face-l...

here is as a live demo from google

https://storage.googleapis.com/tfjs-models/demos/face-landma...

here is a cool example using d3.js to render the earth using Houdini https://observablehq.com/@fil/hello-paintworklet