Code for the paper "Generative Adversarial Imitation Learning"

A living collection of deep learning problems

An educational resource to help anyone learn deep reinforcement learning.

Educational framework exploring ergonomic, lightweight multi-agent orchestration. Managed by OpenAI Solution team.

The OpenAI Model Spec

A multiprotocol and multiplatform quantum random number generation framework

Template for Qiskit python projects

Qiskit is an open-source SDK for working with quantum computers at the level of extended quantum circuits, operators, and primitives.

Zero Noise Extrapolation (ZNE) prototype for error mitigation on the Qiskit Estimator primitive

IBM Quantum Engineering and Enabling Technologies

A subset of PyTorch's neural network modules, written in Python using OpenAI's Triton.

An experiment in open source at the Department of Defense.

Robust Speech Recognition via Large-Scale Weak Supervision

Hackable and optimized Transformers building blocks, supporting a composable construction.

Cramming the training of a (BERT-type) language model into limited compute.

LlamaIndex is the leading document agent and OCR platform

The agent engineering platform.

Tesseract Open Source OCR Engine (main repository)

The TinyLlama project is an open endeavor to pretrain a 1.1B Llama model on 3 trillion tokens.

Unsloth Studio is a web UI for training and running open models like Gemma 4, Qwen3.6, DeepSeek, gpt-oss locally.

Code and documentation to train Stanford's Alpaca models, and generate the data.

Run, manage, and scale AI workloads on any AI infrastructure. Use one system to access & manage all AI compute (Kubernetes, Slurm, 20+ clouds, on-prem).

Vane is an AI-powered answering engine.

Run PyTorch LLMs locally on servers, desktop and mobile

A PyTorch native platform for training generative AI models

Frontier Models playing the board game Diplomacy.

Implementing DeepSeek R1's GRPO algorithm from scratch