Evals is a framework for evaluating LLMs and LLM systems, and an open-source registry of benchmarks.

[CoRL '23] Dexterous piano playing with deep reinforcement learning.

Lightweight wrapper of the official ChatGPT API in your terminal

Train to 94% on CIFAR-10 in <6.3 seconds on a single A100. Or ~95.79% in ~110 seconds (or less!)

The simplest, fastest repository for training/finetuning medium-sized GPTs.

Point cloud diffusion for 3D model synthesis

tiktoken is a fast BPE tokeniser for use with OpenAI's models.

Challenges and Opportunities in Offline Reinforcement Learning from Visual Observations

Chrome Extension that Integrates ChatGPT (Unofficial) into Google Search

Real-time behaviour synthesis with MuJoCo, using Predictive Control

Examples and guides for using the OpenAI API

Perceiver-Actor: A Multi-Task Transformer for Robotic Manipulation

Foundation Architecture for (M)LLMs

Drive a browser with GPT-3

A demonstration of how a toy (but usable!) semantic search engine can be quickly built using Cohere's platform.

jax-triton contains integrations between JAX and OpenAI Triton

Offline RL experiments

This repository contains the implementation of the PTR algorithm described in the paper: Pre-Training for Robots: Leveraging Diverse Multitask Data via Offline Reinforcement Learning.

An API standard for single-agent reinforcement learning environments, with popular reference environments and related utilities (formerly Gym)

Train transformer language models with reinforcement learning.

A standard format for offline reinforcement learning datasets, with popular reference datasets and related utilities

Python-based research interface for blackbox and hyperparameter optimization, based on the internal Google Vizier Service.

Official Implementation of Holo-Dex: Teaching Dexterity with Immersive Mixed Reality

Gym environment for playing Wordle with RL agents

🤗 Transformers: the model-definition framework for state-of-the-art machine learning models in text, vision, audio, and multimodal models, for both inference and training.

A modular RL library to fine-tune language models to human preferences

MiniWoB++: a web interaction benchmark for reinforcement learning