Lyra: A New Very Low-Bitrate Codec for Speech Compression

Lyra, a high-quality, very low-bitrate speech codec that makes voice communication available even on the slowest networks
Read more →

Ceres - an MCTS chess engine for research and recreation

Ceres (“Chess Engine for Research”) is:
Read more →

Upgrade

Set in the near-future, technology controls nearly all aspects of life.
Read more →

AlphaStar: Mastering the Real-Time Strategy Game StarCraft II

Games have been used for decades as an important way to test and evaluate the performance of artificial intelligence systems.
Read more →

Learning Complex Goals with Iterated Amplification

We’re proposing an AI safety technique called iterated amplification that lets us specify complicated behaviors and goals that are beyond human scale, by demonstrating how to decompose a task into simpler sub-tasks, rather than by providing labeled data or a reward function.
Read more →

Training AI agents to solve unfamiliar tasks

Composable planning is a new way of building artificial intelligence (AI) agents that are better at solving unfamiliar tasks.
Read more →

OpenAI Five Benchmark: Results

Yesterday, OpenAI Five won a best-of-three against a team of 99.
Read more →

Capture the Flag: the emergence of complex cooperative agents

Mastering the strategy, tactical understanding, and team play involved in multiplayer video games represents a critical challenge for AI research.
Read more →

OpenAI Five

Our team of five neural networks, OpenAI Five, has started to defeat amateur human teams at Dota 2.
Read more →