sankeerthrao

DeepSpeed

DeepSpeed is a deep learning optimization library that makes distributed training and inference easy, efficient, and effective.

flax

Flax is a neural network library for JAX that is designed for flexibility.

mobile-aloha

Mobile ALOHA: Learning Bimanual Mobile Manipulation with Low-Cost Whole-Body Teleoperation

OLMo

Modeling, training, eval, and inference code for OLMo

OLMo-Eval

Evaluation suite for LLMs

open-instruct

optax

Optax is a gradient processing and optimization library for JAX.

paxml

Pax is a Jax-based machine learning framework for training large scale models. Pax allows for advanced and fully configurable experimentation and parallelization, and has demonstrated industry leading model flop utilization rates.

serl

SERL: A Software Suite for Sample-Efficient Robotic Reinforcement Learning

training-frameworks

Testing out various training frameworks for large language models

visionOS

Playing around with visionOS

alphageometry

DALL-E

PyTorch package for the discrete VAE used for DALL·E.

dm-haiku

JAX-based neural network library

gpt-2

Code for the paper "Language Models are Unsupervised Multitask Learners"

legion

The Legion Parallel Programming System

linux

Linux kernel source tree

LLaVA

[NeurIPS'23 Oral] Visual Instruction Tuning (LLaVA) built towards GPT-4V level capabilities and beyond.

marker

Convert PDF to markdown quickly with high accuracy

model-viewer

Easily display interactive 3D models on the web and in AR!

objax

open_clip

An open source implementation of CLIP.

OpenGrip

a simple, low cost, robotics gripper

ROCm

AMD ROCm™ Software - GitHub Home

scenic

Scenic: A Jax Library for Computer Vision Research and Beyond

surya

OCR and line detection in 90+ languages

TensorRT-LLM

TensorRT-LLM provides users with an easy-to-use Python API to define Large Language Models (LLMs) and build TensorRT engines that contain state-of-the-art optimizations to perform inference efficiently on NVIDIA GPUs. TensorRT-LLM also contains components to create Python and C++ runtimes that execute those TensorRT engines.

text

Making text a first-class citizen in TensorFlow.

TinyGPT-V

TinyGPT-V: Efficient Multimodal Large Language Model via Small Backbones

Video-Pre-Training

Video PreTraining (VPT): Learning to Act by Watching Unlabeled Online Videos