fal

Team
company
Verified
Activity Feed

Inference Provider

VERIFIED
158,926 monthly requests

AI & ML interests

generative media platform for developers

gokaygokayΒ 
posted an update 8 months ago
view post
Post
8757
FlashPack: Lightning-Fast Model Loading for PyTorch

https://github.com/fal-ai/flashpack

FlashPack β€” a new, high-throughput file format and loading mechanism for PyTorch that makes model checkpoint I/O blazingly fast, even on systems without access to GPU Direct Storage (GDS).

With FlashPack, loading any model can be 3–6Γ— faster than with the current state-of-the-art methods like accelerate or the standard load_state_dict() and to() flow β€” all wrapped in a lightweight, pure-Python package that works anywhere.

  • 3 replies
Β·
rizaveliogluΒ 
posted an update about 1 year ago
view post
Post
1045
TryOffDiff goes multi-garment!

We're excited to share TryOffDiff v2, extending our approach to support multiple garment categories. Key updates include:
- Training on the multi-garment DressCode dataset, covering upper-body, lower-body, and dresses.
- A simplified adapter design for improved training efficiency and modularity.
- Introduction of four specialized models:
- One model per category (upper, lower, dress),
- Plus a multi-garment model capable of generating multiple garments sequentially from a single image.

*PS:* Visit us this Friday at 10:30 AM in ExHall-B for our live demo @CVPR '25!

Demo: rizavelioglu/tryoffdiff
Project page: https://rizavelioglu.github.io/tryoffdiff
  • 1 reply
Β·
rizaveliogluΒ 
posted an update over 1 year ago
view post
Post
3809
Comparing reconstruction quality of various VAEs with an interactive demo
rizavelioglu/vae-comparison
  • 1 reply
Β·
benjamin-paineΒ 
posted an update over 1 year ago
view post
Post
3304
Zonos is flying up the trending tab, and for good reason - it's the most expressive and emotive open-source TTS I've used to date. I'm happy to say it's now supported in Taproot, with added long-form synthesis support and other goodies.

Try it here: https://huggingface.co/spaces/benjamin-paine/zonos-longform

Getting started with Zonos in Taproot is easy; with a working CUDA toolkit and Python/Pip installation, all you have to do is:
apt install espeak-ng
pip install taproot
taproot install speech-synthesis:zonos-transformer
taproot invoke speech-synthesis:zonos-transformer --text "Hello, world!"

See more on GitHub at https://github.com/painebenjamin/taproot/
  • 2 replies
Β·
benjamin-paineΒ 
posted an update over 1 year ago
view post
Post
4199
Hello HuggingFace πŸ€—, and happy new year! πŸŽ†

I'm thrilled to be releasing the first iteration of a project I've been working on for quite awhile now. It's called Taproot, and it's a seamlessly scalable open-source AI/ML inference engine designed for letting developers build real-time experiences clustered across a small-to-mid-sized cluster, without the burden of hyperscale infrastructure.

Along with the server and task framework is a client library for node and the browser. And what good is a server and client without an app to go alongside it? To that end, I'm also releasing Anachrovox, a fun, real-time hands-free voice assistant that can run on mid-level devices in <12GB VRAM, with web search, weather, and other tools. It uses my real-time browser wake-word library to detect utterances of the phrase 'Hey Vox', 'Hi Vox', 'Okay Vox', 'Anachrovox' or just 'Vox' (alongside some others.)

Releasing this many things at once will definitely result in bugs, so please report them when sighted! Thank you all!

Taproot: https://github.com/painebenjamin/taproot
Taproot JS Client: https://github.com/painebenjamin/taproot.js
Anachrovox: https://github.com/painebenjamin/anachrovox

The Anachrovox Spaces are networked together, balancing load across them to keep all front-ends responsive. You only have to choose what color you like the most!

https://huggingface.co/spaces/benjamin-paine/anachrovox
https://huggingface.co/spaces/benjamin-paine/anachrovox-amber
  • 12 replies
Β·
rizaveliogluΒ 
posted an update over 1 year ago
view post
Post
2321
Introducing Virtual Try-Off (VTOFF), a novel task focused on generating standardized garment images from single photos of clothed individuals. Unlike traditional Virtual Try-On (VTON), which digitally dresses models, VTOFF aims to extract a canonical garment image, posing unique challenges in capturing garment shape, texture, and intricate patterns.

Try it out: rizavelioglu/tryoffdiff
Paper: TryOffDiff: Virtual-Try-Off via High-Fidelity Garment Reconstruction using Diffusion Models (2411.18350)
Project page: https://rizavelioglu.github.io/tryoffdiff
  • 3 replies
Β·
gokaygokayΒ 
posted an update almost 2 years ago
view post
Post
23821
FLUX Prompt Generator Updates

- gokaygokay/FLUX-Prompt-Generator

- There are now hundreds of new selections across diverse categories, each offering a lot of choices:

Architecture, Art, Artist, Brands, Character, Cinematic, Fashion, Feelings, Geography, Human, Interaction, Keywords, Objects, People, Photography, Plots, Poses, Scene, Science, Stuff, Time, Typography, Vehicle, Video Game

- In addition to Hugging Face, I've integrated new LLM providers: Groq, OpenAI, and Claude.

- Upgraded Vision Language Models (VLMs): We now feature Qwen2-VL, JoyCaption and Florence-2-large.

- New specialized system prompts for various styles and themes, including Happy, Simple, Poster, Only Objects, No Figure, Landscape, Fantasy.
  • 6 replies
Β·
isidenticalΒ 
posted an update almost 2 years ago
isidenticalΒ 
posted an update almost 2 years ago
view post
Post
747
Added FLUX.1 pro/dev/schnell and AuraFlow v0.2 to fal/imgsys !!! Go play with it and get us some votez
isidenticalΒ 
posted an update almost 2 years ago
view post
Post
2021
fal/AuraFlow-v0.3 is now here with support for different aspect resolutions (w/h up to 1536px!) and much nicer aesthetics! Make sure to install the latest diffusers to get support for it.
gokaygokayΒ 
posted an update almost 2 years ago
view post
Post
20963
I've built a space for creating prompts for FLUX

gokaygokay/FLUX-Prompt-Generator

You can create long prompts from images or simple words. Enhance your short prompts with prompt enhancer. You can configure various settings such as artform, photo type, character details, scene details, style, and artist to create tailored prompts.

And you can combine all of them with custom prompts using llms (Mixtral, Mistral, Llama 3, and Mistral-Nemo).

The UI is a bit complex, but it includes almost everything you need. Choosing random option is the most fun!

And i've created some other spaces for using FLUX models with captioners and enhancers.

- gokaygokay/FLUX.1-dev-with-Captioner
  • 4 replies
Β·
gokaygokayΒ 
posted an update almost 2 years ago
gokaygokayΒ 
posted an update almost 2 years ago
isidenticalΒ 
posted an update almost 2 years ago
view post
Post
4098
Announcing the second open model in our Aura series of media models at @fal : fal/AuraFlow

Try it using diffusers or ComfyUI from publicly available weights, and read more about it in our blog https://blog.fal.ai/auraflow.
  • 3 replies
Β·
gokaygokayΒ 
posted an update almost 2 years ago