Google DeepMind

All

311 repositories

torax
Public
TORAX: Tokamak transport simulation in JAX
Python
•
Other
•42•400•36•12•Updated Mar 3, 2025Mar 3, 2025
optax
Public
Optax is a gradient processing and optimization library for JAX.
machine-learning optimization
Python
•
Apache License 2.0
•213•1.8k•43•28•Updated Mar 3, 2025Mar 3, 2025
gemma
Public
Open weights LLM from Google DeepMind.
Python
•
Apache License 2.0
•340•2.6k•9•31•Updated Mar 3, 2025Mar 3, 2025
android_env
Public
RL research on Android devices.
android reinforcement-learning
Python
•
Apache License 2.0
•82•1k•0•8•Updated Mar 3, 2025Mar 3, 2025
materials_discovery
Public
Jupyter Notebook
•
Apache License 2.0
•153•951•21•7•Updated Mar 3, 2025Mar 3, 2025
fusion_transport_surrogates
Public
A library of surrogate transport models for tokamak fusion.
Python
•
Apache License 2.0
•0•3•0•1•Updated Mar 3, 2025Mar 3, 2025
mujoco
Public
Multi-Joint dynamics with Contact. A general purpose physics simulator.
robotics physics mujoco
Jupyter Notebook
•
Apache License 2.0
•907•8.8k•99•37•Updated Mar 3, 2025Mar 3, 2025
implicit_diffusion
Public
Python
•
Apache License 2.0
•0•0•0•0•Updated Mar 3, 2025Mar 3, 2025
concordia
Public
A library for generative social simulation
multi-agent social-simulation agent-based-simulation generative-agents
Python
•
Apache License 2.0
•174•791•7•2•Updated Mar 2, 2025Mar 2, 2025
bbeh
Public
Apache License 2.0
•2•32•2•1•Updated Mar 2, 2025Mar 2, 2025
meltingpot
Public
A suite of test scenarios for multi-agent reinforcement learning.
multiagent-reinforcement-learning
Python
•
Apache License 2.0
•128•655•7•2•Updated Mar 1, 2025Mar 1, 2025
century
Public
Apache License 2.0
•0•0•0•0•Updated Feb 28, 2025Feb 28, 2025
onetwo
Public
Python
•
Apache License 2.0
•14•188•2•1•Updated Feb 28, 2025Feb 28, 2025
iris
Public
Python
•
Apache License 2.0
•4•9•0•2•Updated Feb 28, 2025Feb 28, 2025
physics-IQ-benchmark
Public
Benchmarking physical understanding in generative video models
benchmark generative-models video-generation physical-understanding
Python
•
Other
•12•118•2•0•Updated Feb 28, 2025Feb 28, 2025
dm_control
Public
Google DeepMind's software stack for physics-based simulation and Reinforcement Learning environments, using MuJoCo.
machine-learning reinforcement-learning deep-learning artificial-intelligence neural-networks physics-simulation mujoco
Python
•
Apache License 2.0
•685•4k•97•8•Updated Feb 28, 2025Feb 28, 2025
kfac-jax
Public
Second Order Optimization and Curvature Estimation with K-FAC in JAX.
machine-learning optimization bayesian-deep-learning
Python
•
Apache License 2.0
•25•264•10•7•Updated Feb 27, 2025Feb 27, 2025
open_spiel
Public
OpenSpiel is a collection of environments and algorithms for research in general reinforcement learning and search/planning in games.
cpp multiagent python games reinforcement-learning
C++
•
Apache License 2.0
•963•4.4k•23•12•Updated Feb 27, 2025Feb 27, 2025
md4
Public
Official Jax Implementation of MD4 Masked Diffusion Models
Python
•
Apache License 2.0
•6•62•0•0•Updated Feb 27, 2025Feb 27, 2025
distrax
Public
Python
•
Apache License 2.0
•32•549•22•20•Updated Feb 27, 2025Feb 27, 2025
recoglab
Public
Python
•
Apache License 2.0
•0•1•0•0•Updated Feb 26, 2025Feb 26, 2025
autonumerics_zero
Public
Python
•
Apache License 2.0
•0•2•0•0•Updated Feb 26, 2025Feb 26, 2025
tapnet
Public
Tracking Any Point (TAP)
benchmark computer-vision deep-learning robotics point-tracking
Jupyter Notebook
•
Apache License 2.0
•137•1.4k•25•1•Updated Feb 25, 2025Feb 25, 2025
chex
Public
Python
•
Apache License 2.0
•47•827•21•20•Updated Feb 25, 2025Feb 25, 2025
loft
Public
LOFT: A 1 Million+ Token Long-Context Benchmark
Python
•
Apache License 2.0
•13•174•2•2•Updated Feb 24, 2025Feb 24, 2025
alphafold3
Public
AlphaFold 3 inference pipeline.
Python
•
Other
•759•6.1k•14•0•Updated Feb 24, 2025Feb 24, 2025
mujoco_playground
Public
An open-source library for GPU-accelerated robot learning and sim-to-real transfer.
Jupyter Notebook
•
Apache License 2.0
•70•746•15•0•Updated Feb 24, 2025Feb 24, 2025
mujoco_menagerie
Public
A collection of high-quality models for the MuJoCo physics engine, curated by Google DeepMind.
robotics mujoco
Python
•
Other
•259•1.8k•10•7•Updated Feb 24, 2025Feb 24, 2025
simulation_streams
Public
Simulation Streams is a programming paradigm designed to efficiently control and leverage Large Language Models (LLMs) for complex, dynamic simulations and agentic workflows.
ai simulations agents llms
Python
•
Apache License 2.0
•2•14•0•0•Updated Feb 21, 2025Feb 21, 2025
recurrentgemma
Public
Open weights language model from Google DeepMind, based on Griffin.
Python
•
Apache License 2.0
•30•623•0•1•Updated Feb 20, 2025Feb 20, 2025