Pegada pública no GitHub de Google DeepMind

Multi-Joint dynamics with Contact. A general purpose physics simulator.

This repository contains implementations and illustrative code to accompany DeepMind publications

Open source code for AlphaFold 2.

TensorFlow-based neural network library

AlphaFold 3 inference pipeline.

StarCraft II Learning Environment

Nenhuma descrição fornecida para este repositório.

A customisable 3D platform for agent-based AI research

Gemma open-weight LLM library, from Google DeepMind

OpenSpiel is a collection of environments and algorithms for research in general reinforcement learning and search/planning in games.

Build Graph Nets in Tensorflow

Nenhuma descrição fornecida para este repositório.

Google DeepMind's software stack for physics-based simulation and Reinforcement Learning environments, using MuJoCo.

A collection of high-quality models for the MuJoCo physics engine, curated by Google DeepMind.

A library of reinforcement learning components and agents

GDM Science Skills to speed up agentic scientific workflows with better grounding and higher token efficiency. Integrate insights from AlphaGenome, AFDB, UniProt and 30+ other databases and tools.

JAX-based neural network library

Nenhuma descrição fornecida para este repositório.

Monte Carlo tree search in JAX

Optax is a gradient processing and optimization library for JAX.

An open-source library for sim and real robot demos on top of MuJoCo.

Tracking Any Point (TAP)

This API provides programmatic access to the AlphaGenome model developed by Google DeepMind.

Nenhuma descrição fornecida para este repositório.

This dataset code generates mathematical question and answer pairs, from a range of question types at roughly school-level difficulty.

A library for generative social simulation

Real-time behaviour synthesis with MuJoCo, using Predictive Control

GPU-optimized version of the MuJoCo physics simulator, designed for NVIDIA hardware.

Nenhuma descrição fornecida para este repositório.

A collection of formalized statements of conjectures in Lean.

RL research on Android devices.

Nenhuma descrição fornecida para este repositório.

Nenhuma descrição fornecida para este repositório.

Dramatron uses large language models to generate coherent scripts and screenplays.

Nenhuma descrição fornecida para este repositório.

Nenhuma descrição fornecida para este repositório.

A platform for managing machine learning experiments

Research code accompanying AlphaGenome

A suite of test scenarios for multi-agent reinforcement learning.

An implementation of the Fermionic Neural Network for ab-initio electronic structure calculations

Nenhuma descrição fornecida para este repositório.

Reverb is an efficient and easy-to-use data storage and transport system designed for machine learning research

official code for "Large Language Models as Optimizers"

Accompanying code for "Discovering State-of-the-art Reinforcement Algorithms" Nature publication

TORAX: Tokamak transport simulation in JAX

Benchmarking long-form factuality in large language models. Original code for our paper "Long-form factuality in large language models".

Open weights language model from Google DeepMind, based on Griffin.

On the Theoretical Limitations of Embedding-Based Retrieval

hanabi_learning_environment is a research platform for Hanabi experiments.

Grandmaster-Level Chess Without Search

TIPSv2 (CVPR'26) and TIPS (ICLR'25)

Lua/Torch implementation of DQN (Nature, 2015)

This is a suite of reinforcement learning environments illustrating various safety properties of intelligent agents.

Nenhuma descrição fornecida para este repositório.

Nenhuma descrição fornecida para este repositório.

DQN Zoo is a collection of reference implementations of reinforcement learning agents developed at DeepMind based on the Deep Q-Network (DQN) agent.

PIX is an image processing library in JAX, for JAX.

Foundation models for 4D spatial and temporal vision tasks.

Official repository for "VideoPrism: A Foundational Visual Encoder for Video Understanding" (ICML 2024)

Benchmarking physical understanding in generative video models

Barkour Robot: Agile Quadruped Robots by Google DeepMind

Spriteworld: a flexible, configurable python-based reinforcement learning environment

Library for sequence-to-sequence numeric prediction, applicable to any tokenizable input, and allows pretraining and fine-tuning over multiple tasks.

Second Order Optimization and Curvature Estimation with K-FAC in JAX.

Nenhuma descrição fornecida para este repositório.

AlphaProtein Novo

Nenhuma descrição fornecida para este repositório.

Nenhuma descrição fornecida para este repositório.

Nenhuma descrição fornecida para este repositório.

A code generation dataset for generating the code that implements Hearthstone and Magic The Gathering card effects.

Nenhuma descrição fornecida para este repositório.

Nenhuma descrição fornecida para este repositório.

Nenhuma descrição fornecida para este repositório.

Nenhuma descrição fornecida para este repositório.

Jeo: Jax model training lib for Earth Observation

Loopy belief propagation for factor graphs on discrete variables in JAX

Official Jax Implementation of MD4 Masked Diffusion Models

GeeFlow - generate and process large-scale geospatial datasets with Google Earth Engine.

Nenhuma descrição fornecida para este repositório.

Datasets to protect Earth's forests and biodiversity

AlphaGeometry2 symbolic engine (DDAR) with examples

A JAX Research Toolkit for Visualizing, Manipulating, and Understanding Gemma Models with Multi-modal Support based on Penzai.

Amplio: A Lightweight Agent Harness for Robust and Long-Horizon Runs

This repository contains levels for boxoban, a box-pushing puzzle game inspired by Sokoban.

Nenhuma descrição fornecida para este repositório.

Some Cryptids and Lean proofs

Nenhuma descrição fornecida para este repositório.

Nenhuma descrição fornecida para este repositório.

Nenhuma descrição fornecida para este repositório.

Build AI text detectors whose decisions are mechanistically grounded and human-auditable (EMNLP 2026)

SynthID Bio is a family of methods developed by Google DeepMind for embedding highly detectable yet function-preserving watermarks directly into AI-generated biological sequences and structures.

PhaseCoder: Microphone Geometry-Agnostic Spatial Audio Understanding for Multimodal LLMs

Nenhuma descrição fornecida para este repositório.

Lean formalizations of Putnam-like problems

"Making group conversations more accessible. Real-time speaker localization and directional guidance for mobile speech-to-text. ACM CHI 2025 Best Paper Award."

Nenhuma descrição fornecida para este repositório.

simulation framework for real robot workflows

Nenhuma descrição fornecida para este repositório.

Nenhuma descrição fornecida para este repositório.

Implementation of unPIC from "How to Spin an Object: First, Get the Shape Right", a CVPRW 2026 paper.