Brand new TTS solution

Huggingface deployment for FastHTML

dstack is an open-source alternative to Kubernetes, designed to simplify development, training, and deployment of AI across any cloud or on-prem. It supports NVIDIA, AMD, and TPU.

The AI Scientist: Towards Fully Automated Open-Ended Scientific Discovery 🧑‍🔬

Official implementation of OV-DINO: Unified Open-Vocabulary Detection with Language-Aware Selective Fusion

Simple tools to mix and match PyTorch and Jax - Get the best of both worlds!

The official Meta Llama 3 GitHub site

LLM training in simple, raw C/CUDA

Real-time audio to chords, lyrics, beat, and melody.

🙌 OpenHands: Code Less, Make More

From scratch implementation of a sparse mixture of experts language model inspired by Andrej Karpathy's makemore :)

Code, Dataset, and Pretrained Models for Audio and Speech Large Language Model "Listen, Think, and Understand".

Finetune Llama 3.1, Mistral, Phi & Gemma LLMs 2-5x faster with 80% less memory

Build reactive web UIs in Python

Minimal, clean code for the Byte Pair Encoding (BPE) algorithm commonly used in LLM tokenization.

Official Code for Stable Cascade

Foundational model for human-like, expressive TTS

Contrastive Language-Audio Pretraining

Local LLM-based social network filter

WhisperFusion builds upon the capabilities of WhisperLive and WhisperSpeech to provide a seamless conversations with an AI.

Simple, Elegant, Typed Argument Parsing with argparse

Karras et al. (2022) diffusion models for PyTorch

A toy programming language with time travel

don't ask

All things prompt engineering

A powerful set of Python debugging tools, based on PySnooper

Convert PDF to markdown quickly with high accuracy

Continual pretraining of foundation LLM using ⚡ Lightning Fabric

Video-LLaVA: Learning United Visual Representation by Alignment Before Projection

Absolute minimalistic implementation of a GPT-like transformer using only numpy (<650 lines).