openai-cookbookExamples and guides for using the OpenAI API
whisperRobust Speech Recognition via Large-Scale Weak Supervision
gymA toolkit for developing and comparing reinforcement learning algorithms.
gpt-2Code for the paper "Language Models are Unsupervised Multitask Learners"
chatgpt-retrieval-pluginThe ChatGPT Retrieval Plugin lets you easily find personal or work documents by asking questions in natural language.
CLIPCLIP (Contrastive Language-Image Pretraining), Predict the most relevant text snippet given an image
gpt-3GPT-3: Language Models are Few-Shot Learners
openai-pythonThe official Python library for the OpenAI API
baselinesOpenAI Baselines: high-quality implementations of reinforcement learning algorithms
evalsEvals is a framework for evaluating LLMs and LLM systems, and an open-source registry of benchmarks.
DALL-EPyTorch package for the discrete VAE used for DALL·E.
shap-eGenerate 3D objects conditioned on text or images
tritonDevelopment repository for the Triton language and compiler
spinningupAn educational resource to help anyone learn deep reinforcement learning.
tiktokentiktoken is a fast BPE tokeniser for use with OpenAI's models.
universeUniverse: a software platform for measuring and training an AI's general intelligence across the world's supply of games, websites and other applications.
jukeboxCode for the paper "Jukebox: A Generative Model for Music"
point-ePoint cloud diffusion for 3D model synthesis
consistency_modelsOfficial repo for consistency models.
openai-nodeThe official Node.js / Typescript library for the OpenAI API
plugins-quickstartGet a ChatGPT plugin up and running in under 5 minutes!
glide-text2imGLIDE: a diffusion-based text-conditional image synthesis model
retroRetro Games in Gym
glowCode for reproducing results in "Glow: Generative Flow with Invertible 1x1 Convolutions"
mujoco-pyMuJoCo is a physics engine for detailed, efficient rigid body simulations with contacts. mujoco-py allows using MuJoCo from Python 3.
openai-quickstart-nodeNode.js example app from the OpenAI API quickstart tutorial
improved-ganCode for the paper "Improved Techniques for Training GANs"
improved-diffusionRelease for Improved Denoising Diffusion Probabilistic Models
roboschoolDEPRECATED: Open-source software for robot simulation, integrated with OpenAI Gym.
finetune-transformer-lmCode and model for the paper "Improving Language Understanding by Generative Pre-Training"
multiagent-particle-envsCode for a multi-agent particle environment used in the paper "Multi-Agent Actor-Critic for Mixed Cooperative-Competitive Environments"
pixel-cnnCode for the paper "PixelCNN++: A PixelCNN Implementation with Discretized Logistic Mixture Likelihood and Other Modifications"
gpt-2-output-datasetDataset of GPT-2 outputs for research in detection, biases, and more
requests-for-researchA living collection of deep learning problems
gpt-discord-botExample Discord bot written in Python that uses the completions API to have conversations with the `text-davinci-003` model, and the moderations API to filter the messages.
human-evalCode for the paper "Evaluating Large Language Models Trained on Code"
multi-agent-emergence-environmentsEnvironment generation code for the paper "Emergent Tool Use From Multi-Agent Autocurricula"
openai-quickstart-pythonPython example app from the OpenAI API quickstart tutorial
evolution-strategies-starterCode for the paper "Evolution Strategies as a Scalable Alternative to Reinforcement Learning"
generating-reviews-discovering-sentimentCode for "Learning to Generate Reviews and Discovering Sentiment"
neural-mmoCode for the paper "Neural MMO: A Massively Multiagent Game Environment for Training and Evaluating Intelligent Agents"
sparse_attentionExamples of using sparse attention, as in "Generating Long Sequences with Sparse Transformers"
maddpgCode for the MADDPG algorithm from the paper "Multi-Agent Actor-Critic for Mixed Cooperative-Competitive Environments"
universe-starter-agentA starter agent that can solve a number of universe environments.
Video-Pre-TrainingVideo PreTraining (VPT): Learning to Act by Watching Unlabeled Online Videos
InfoGANCode for reproducing key results in the paper "InfoGAN: Interpretable Representation Learning by Information Maximizing Generative Adversarial Nets"
supervised-reptileCode for the paper "On First-Order Meta-Learning Algorithms"
prm800k800,000 step-level correctness labels on LLM solutions to MATH problems
blocksparseEfficient GPU kernels for block-sparse matrix multiplication and convolution
procgenProcgen Benchmark: Procedurally-Generated Game-Like Gym-Environments
lm-human-preferencesCode for the paper Fine-Tuning Language Models from Human Preferences
kubernetes-ec2-autoscalerA batch-optimized scaling manager for Kubernetes
summarize-from-feedbackCode for "Learning to summarize from human feedback"
random-network-distillationCode for the paper "Exploration by Random Network Distillation"
large-scale-curiosityCode for the paper "Large-Scale Study of Curiosity-Driven Learning"
multiagent-competitionCode for the paper "Emergent Complexity via Multi-agent Competition"
imitationCode for the paper "Generative Adversarial Imitation Learning"
deeptypeCode for the paper "DeepType: Multilingual Entity Linking by Neural Type System Evolution"
mlshCode for the paper "Meta-Learning Shared Hierarchies"
openai-openapiOpenAPI specification for the OpenAI API
iafCode for reproducing key results in the paper "Improving Variational Inference with Inverse Autoregressive Flow"
mujoco-worldgenAutomatic object XML generation for Mujoco
safety-gymTools for accelerating safe exploration research.
vdvaeRepository for the paper "Very Deep VAEs Generalize Autoregressive Models and Can Outperform Them on Images"
coinrunCode for the paper "Quantifying Transfer in Reinforcement Learning"
weightnormExample code for Weight Normalization, from "Weight Normalization: A Simple Reparameterization to Accelerate Training of Deep Neural Networks"
atari-pyA packaged and slightly-modified version of https://github.com/bbitmaster/ale_python_interface
robogymRobotics Gym Environments
openai-gemmOpen single and half precision gemm implementations
vimeCode for the paper "Curiosity-driven Exploration in Deep Reinforcement Learning via Bayesian Neural Networks"
safety-starter-agentsBasic constrained RL agents used in experiments for the "Benchmarking Safe Exploration in Deep Reinforcement Learning" paper.
ebm_code_releaseCode for Implicit Generation and Generalization with Energy Based Models
CLIP-featureviscode for reproducing some of the diagrams in the paper "Multimodal Neurons in Artificial Neural Networks"
gym-http-apiAPI to access OpenAI Gym from other languages via HTTP
robosumoCode for the paper "Continuous Adaptation via Meta-Learning in Nonstationary and Competitive Environments"
EPGCode for the paper "Evolved Policy Gradients"
orrbCode for the paper "OpenAI Remote Rendering Backend"
phasic-policy-gradientCode for the paper "Phasic Policy Gradient"
miniF2FFormal to Formal Mathematics Benchmark
atari-resetCode for the blog post "Learning Montezuma’s Revenge from a Single Demonstration"
spinningup-workshopFor educational materials related to the spinning up workshops.
train-procgenCode for the paper "Leveraging Procedural Generation to Benchmark Reinforcement Learning"
dallify-discord-botExample code for using OpenAI’s NodeJS SDK with discord.js SDK to create a Discord Bot that uses Slash Commands.
gym3Vectorized interface for reinforcement learning environments
retro-baselinesPublicly releasable baselines for the Retro contest
neural-gpuCode for the Neural GPU model originally described in "Neural GPUs Learn Algorithms"
go-vncdriverFast VNC driver
human-eval-infillingCode for the paper "Efficient Training of Language Models to Fill in the Middle"
tabulatepublic release of Excel / OpenAI API integration
distribution_augmentationCode for the paper, "Distribution Augmentation for Generative Modeling", ICML 2020.
consistency_models_cifar10Consistency models trained on CIFAR-10, in JAX.