Releases

Google Research introduces Imagen, a text-to-image diffusion model emphasizing photorealism and language understanding.

Google Research introduces the 540-billion-parameter Pathways Language Model and reports strong few-shot performance across hundreds of tasks.

DeepMind and EMBL-EBI launch a freely accessible database of AlphaFold protein structure predictions.

DeepMind publishes AlphaFold 2's methodology and releases source code, expanding access to high-accuracy protein prediction.

Google announces fourth-generation TPU pods for large-scale machine-learning training.

Google Cloud launches Vertex AI as a unified platform for building, deploying and operating ML models.

Google introduces LaMDA as a conversational language model designed for open-ended dialogue.

DeepMind publishes MuZero, which plans using a learned model without being told the environment's underlying rules.

AlphaFold 2 demonstrates accuracy competitive with experimental structures at CASP14, marking a major protein-folding breakthrough.

Google researchers publish the T5 approach and Colossal Clean Crawled Corpus, framing diverse NLP tasks in one text-to-text format.

Google releases MediaPipe as a cross-platform framework for building perception and multimodal ML pipelines.

DeepMind's first AlphaFold system achieves top performance in the CASP13 protein-structure prediction assessment.

Google Research releases JAX for composable automatic differentiation and accelerator-backed numerical computing.

Google open-sources BERT, enabling bidirectional pre-training for language understanding and downstream NLP tasks.

DeepMind publishes AlphaZero, a general self-play reinforcement-learning system for multiple board games.

Google researchers introduce the Transformer architecture, replacing recurrence with attention and reshaping modern sequence modeling.

At Google I/O, the company frames its strategic transition from mobile-first to AI-first products and computing.

Google announces its second-generation TPU and plans to offer Cloud TPUs to Google Cloud customers.

DeepMind introduces WaveNet, a neural generative model for raw audio waveforms that substantially improves speech synthesis.

Google publicly reveals the TPU, a custom accelerator designed to speed machine-learning workloads in its data centers.

AlphaGo wins its landmark five-game match against 18-time world champion Lee Sedol by four games to one.

Google releases TensorFlow as an open-source machine-learning framework, making internal deep-learning infrastructure broadly available.

AlphaGo becomes the first computer program to defeat a professional Go player without handicaps on a full-sized board.

Google completes its reorganization under Alphabet Inc.; Google becomes the principal operating subsidiary while Alphabet becomes the publicly traded parent.

Google agrees to acquire London-based artificial-intelligence company DeepMind Technologies, bringing its research team into Google.