- San Francisco, CA
- http://tarzain.com
Stars
(CVPR 2025) From Slow Bidirectional to Fast Autoregressive Video Diffusion Models
FastVideo is a unified framework for accelerated video generation.
Official code for "F5-TTS: A Fairytaler that Fakes Fluent and Faithful Speech with Flow Matching"
SpeechGPT Series: Speech Large Language Models
A fast, light, open chat UI with full tool use support across many models
Absolute minimalistic implementation of a GPT-like transformer using only numpy (<650 lines).
A lightweight wrapper around https://github.com/facebookresearch/encodec that enables dynamic streamed reading, seeking, metadata and GPU support.
Handwriting Synthesis with RNNs ✏️
This is the official code release for Bayesian Flow Networks.
A natural language interface for computers
[CVPR 2024] 4D Gaussian Splatting for Real-Time Dynamic Scene Rendering
The TinyLlama project is an open endeavor to pretrain a 1.1B Llama model on 3 trillion tokens.
Stable bi-directional infinite scroll React component
Estimating the COVID risk of ordinary activities
A simple plug and play feedback component for your website.
Get metadata about the active window and open windows (title, id, bounds, owner, etc)
Terraform modules that help you explore Okta and AWS Session Manager integrations
a library to create multi device experiments
Run Keras models in the browser, with GPU support using WebGL
A batch-optimized scaling manager for Kubernetes