Models News
AI models are trained neural network systems — like GPT, Claude and Gemini — that power modern artificial intelligence. This page tracks every model release, benchmark result and capability update as it happens.
2,782 articles · updated 24/7 from 100+ sources
Competitive self-play
We’ve found that self-play allows simulated AIs to discover physical skills like tackling, ducking, faking,…
Nonlinear computation in deep linear networks
Nonlinear computation in deep linear networks. OpenAI News - artificial intelligence news.
Learning to model other minds
We’re releasing an algorithm which accounts for the fact that other agents are learning too, and discovers…
Learning with opponent-learning awareness
Learning with opponent-learning awareness. OpenAI News - artificial intelligence news.
OpenAI Baselines: ACKTR & A2C
We’re releasing two new OpenAI Baselines implementations: ACKTR and A2C. A2C is a synchronous,…
More on Dota 2
Our Dota 2 result shows that self-play can catapult the performance of machine learning systems from far…
Gathering human feedback
RL-Teacher is an open-source implementation of our interface to train AIs via occasional human feedback…
Better exploration with parameter noise
We’ve found that adding adaptive noise to the parameters of reinforcement learning algorithms frequently…
Proximal Policy Optimization
We’re releasing a new class of reinforcement learning algorithms, Proximal Policy Optimization (PPO), which…
Robust adversarial inputs
We’ve created images that reliably fool neural network classifiers when viewed from varied scales and…
Hindsight Experience Replay
Hindsight Experience Replay. OpenAI News - artificial intelligence news.
Teacher–student curriculum learning
Teacher–student curriculum learning. OpenAI News - artificial intelligence news.
Faster physics in Python
We’re open-sourcing a high-performance Python library for robotic simulation using the MuJoCo engine,…
Learning from human preferences
One step towards building safe AI systems is to remove the need for humans to write goal functions, since…
Learning to cooperate, compete, and communicate
Multiagent environments where agents compete for resources are stepping stones on the path to AGI. Multiagent…
UCB exploration via Q-ensembles
UCB exploration via Q-ensembles. OpenAI News - artificial intelligence news.
OpenAI Baselines: DQN
We’re open-sourcing OpenAI Baselines, our internal effort to reproduce reinforcement learning algorithms…
Robots that learn
We’ve created a robotics system, trained entirely in simulation and deployed on a physical robot, which can…
Roboschool
We are releasing Roboschool: open-source software for robot simulation, integrated with OpenAI Gym.
Equivalence between policy gradients and soft Q-learning
Equivalence between policy gradients and soft Q-learning. OpenAI News - artificial intelligence news.
Team update
The OpenAI team is now 45 people. Together, we’re pushing the frontier of AI capabilities—whether by…
OpenAI Gym Beta
We’re releasing the public beta of OpenAI Gym, a toolkit for developing and comparing reinforcement…