# google-deepmind — Python repositories (exhaustive catalog)

Auto-generated from the GitHub Search API (`org:google-deepmind language:python`). Source of truth: `docs/reference/ai/data/google-deepmind-repos.json`.

Total: **236** repos — **205** active, **31** archived. Sorted by stars.

| # | Repo | Stars | Status | Last push | Description |
|--:|------|------:|--------|-----------|-------------|
| 1 | [alphafold](https://github.com/google-deepmind/alphafold) | 14605 | active | 22/04/2 | Open source code for AlphaFold 2. |
| 2 | [sonnet](https://github.com/google-deepmind/sonnet) | 9918 | active | 06/05/2 | TensorFlow-based neural network library |
| 3 | [pysc2](https://github.com/google-deepmind/pysc2) | 8292 | active | 23/07/2 | StarCraft II Learning Environment |
| 4 | [alphafold3](https://github.com/google-deepmind/alphafold3) | 8054 | active | 19/05/2 | AlphaFold 3 inference pipeline. |
| 5 | [graphcast](https://github.com/google-deepmind/graphcast) | 6661 | active | 30/03/2 | - |
| 6 | [graph_nets](https://github.com/google-deepmind/graph_nets) | 5401 | active | 12/12/2 | Build Graph Nets in Tensorflow |
| 7 | [gemma](https://github.com/google-deepmind/gemma) | 5239 | active | 20/05/2 | Gemma open-weight LLM library, from Google DeepMind |
| 8 | [alphageometry](https://github.com/google-deepmind/alphageometry) | 4841 | active | 13/01/2 | - |
| 9 | [dm_control](https://github.com/google-deepmind/dm_control) | 4586 | active | 17/05/2 | Google DeepMind's software stack for physics-based simulation and Reinforcement Learning environments, using MuJoCo. |
| 10 | [learning-to-learn](https://github.com/google-deepmind/learning-to-learn) | 4070 | active | 29/06/2 | Learning to Learn in TensorFlow |
| 11 | [acme](https://github.com/google-deepmind/acme) | 3986 | active | 08/04/2 | A library of reinforcement learning components and agents |
| 12 | [mujoco_menagerie](https://github.com/google-deepmind/mujoco_menagerie) | 3485 | active | 20/05/2 | A collection of high-quality models for the MuJoCo physics engine, curated by Google DeepMind. |
| 13 | [dm-haiku](https://github.com/google-deepmind/dm-haiku) | 3233 | active | 12/05/2 | JAX-based neural network library |
| 14 | [trfl](https://github.com/google-deepmind/trfl) | 3134 | active | 08/12/2 | TensorFlow Reinforcement Learning |
| 15 | [alphatensor](https://github.com/google-deepmind/alphatensor) | 2838 | active | 22/04/2 | - |
| 16 | [mctx](https://github.com/google-deepmind/mctx) | 2626 | active | 02/09/2 | Monte Carlo tree search in JAX |
| 17 | [dnc](https://github.com/google-deepmind/dnc) | 2540 | active | 23/07/2 | A TensorFlow implementation of the Differentiable Neural Computer. |
| 18 | [optax](https://github.com/google-deepmind/optax) | 2269 | active | 18/05/2 | Optax is a gradient processing and optimization library for JAX. |
| 19 | [mathematics_dataset](https://github.com/google-deepmind/mathematics_dataset) | 1955 | active | 23/12/2 | This dataset code generates mathematical question and answer pairs, from a range of question types at roughly school-level difficulty. |
| 20 | [mujoco_playground](https://github.com/google-deepmind/mujoco_playground) | 1954 | active | 19/05/2 | An open-source library for GPU-accelerated robot learning and sim-to-real transfer. |
| 21 | [alphagenome](https://github.com/google-deepmind/alphagenome) | 1924 | active | 06/05/2 | This API provides programmatic access to the AlphaGenome model developed by Google DeepMind. |
| 22 | [penzai](https://github.com/google-deepmind/penzai) | 1884 | active | 22/06/2 | A JAX research toolkit for building, editing, and visualizing neural networks. |
| 23 | [kinetics-i3d](https://github.com/google-deepmind/kinetics-i3d) | 1834 | active | 12/09/2 | Convolutional neural network model for video classification trained on the Kinetics dataset. |
| 24 | [bsuite](https://github.com/google-deepmind/bsuite) | 1544 | active | 30/03/2 | bsuite is a collection of carefully-designed experiments that investigate core capabilities of a reinforcement learning (RL) agent |
| 25 | [jraph](https://github.com/google-deepmind/jraph) | 1470 | `archived` | 18/03/2 | A Graph Neural Network Library in Jax |
| 26 | [concordia](https://github.com/google-deepmind/concordia) | 1433 | active | 06/05/2 | A library for generative social simulation |
| 27 | [rlax](https://github.com/google-deepmind/rlax) | 1426 | active | 31/03/2 | - |
| 28 | [rc-data](https://github.com/google-deepmind/rc-data) | 1296 | `archived` | 26/04/2 | Question answering dataset featured in "Teaching Machines to Read and Comprehend |
| 29 | [mujoco_warp](https://github.com/google-deepmind/mujoco_warp) | 1240 | active | 21/05/2 | GPU-optimized version of the MuJoCo physics simulator, designed for NVIDIA hardware. |
| 30 | [android_env](https://github.com/google-deepmind/android_env) | 1219 | active | 21/05/2 | RL research on Android devices. |
| 31 | [scalable_agent](https://github.com/google-deepmind/scalable_agent) | 1025 | active | 13/03/2 | A TensorFlow implementation of Scalable Distributed Deep-RL with Importance Weighted Actor-Learner Architectures. |
| 32 | [tree](https://github.com/google-deepmind/tree) | 1022 | active | 31/03/2 | tree is a library for working with nested data structures |
| 33 | [chex](https://github.com/google-deepmind/chex) | 941 | active | 10/04/2 | - |
| 34 | [xmanager](https://github.com/google-deepmind/xmanager) | 909 | active | 18/05/2 | A platform for managing machine learning experiments |
| 35 | [synthid-text](https://github.com/google-deepmind/synthid-text) | 867 | active | 06/08/2 | - |
| 36 | [ferminet](https://github.com/google-deepmind/ferminet) | 837 | active | 21/05/2 | An implementation of the Fermionic Neural Network for ab-initio electronic structure calculations |
| 37 | [meltingpot](https://github.com/google-deepmind/meltingpot) | 836 | active | 13/05/2 | A suite of test scenarios for multi-agent reinforcement learning. |
| 38 | [alphagenome_research](https://github.com/google-deepmind/alphagenome_research) | 763 | active | 15/05/2 | Research code accompanying AlphaGenome |
| 39 | [opro](https://github.com/google-deepmind/opro) | 745 | active | 04/12/2 | official code for "Large Language Models as Optimizers" |
| 40 | [alphadev](https://github.com/google-deepmind/alphadev) | 738 | `archived` | 20/06/2 | - |
| 41 | [disco_rl](https://github.com/google-deepmind/disco_rl) | 701 | active | 02/12/2 | Accompanying code for "Discovering State-of-the-art Reinforcement Algorithms" Nature publication |
| 42 | [long-form-factuality](https://github.com/google-deepmind/long-form-factuality) | 685 | active | 20/05/2 | Benchmarking long-form factuality in large language models. Original code for our paper "Long-form factuality in large language models". |
| 43 | [uncertain_ground_truth](https://github.com/google-deepmind/uncertain_ground_truth) | 681 | `archived` | 28/03/2 | Dermatology ddx dataset, Jax implementations of Monte Carlo conformal prediction, plausibility regions and statistical annotation aggregation from our recent work on uncertain ground truth (TMLR'23 and ArXiv pre-print). |
| 44 | [recurrentgemma](https://github.com/google-deepmind/recurrentgemma) | 677 | active | 06/02/2 | Open weights language model from Google DeepMind, based on Griffin. |
| 45 | [torax](https://github.com/google-deepmind/torax) | 673 | active | 18/05/2 | TORAX: Tokamak transport simulation in JAX |
| 46 | [hanabi-learning-environment](https://github.com/google-deepmind/hanabi-learning-environment) | 668 | `archived` | 14/02/2 | hanabi_learning_environment is a research platform for Hanabi experiments. |
| 47 | [pycolab](https://github.com/google-deepmind/pycolab) | 664 | active | 06/09/2 | A highly-customisable gridworld game engine with some batteries included. Make your own gridworld games to test reinforcement learning agents! |
| 48 | [distrax](https://github.com/google-deepmind/distrax) | 634 | active | 13/05/2 | - |
| 49 | [alphamissense](https://github.com/google-deepmind/alphamissense) | 633 | `archived` | 13/03/2 | - |
| 50 | [ai-safety-gridworlds](https://github.com/google-deepmind/ai-safety-gridworlds) | 633 | `archived` | 18/05/2 | This is a suite of reinforcement learning environments illustrating various safety properties of intelligent agents. |
| 51 | [searchless_chess](https://github.com/google-deepmind/searchless_chess) | 631 | active | 10/01/2 | Grandmaster-Level Chess Without Search |
| 52 | [surface-distance](https://github.com/google-deepmind/surface-distance) | 597 | active | 19/02/2 | Library to compute surface distance based performance metrics for segmentation tasks. |
| 53 | [gemini-robotics-sdk](https://github.com/google-deepmind/gemini-robotics-sdk) | 582 | active | 19/05/2 | - |
| 54 | [alphastar](https://github.com/google-deepmind/alphastar) | 567 | active | 08/09/2 | - |
| 55 | [tracr](https://github.com/google-deepmind/tracr) | 565 | `archived` | 05/02/2 | - |
| 56 | [simply](https://github.com/google-deepmind/simply) | 526 | active | 19/05/2 | Minimal and scalable research codebase in JAX, designed for rapid iteration on frontier research in LLM and other autoregressive models. |
| 57 | [dqn_zoo](https://github.com/google-deepmind/dqn_zoo) | 501 | active | 04/05/2 | DQN Zoo is a collection of reference implementations of reinforcement learning agents developed at DeepMind based on the Deep Q-Network (DQN) agent. |
| 58 | [treescope](https://github.com/google-deepmind/treescope) | 470 | active | 08/08/2 | An interactive HTML pretty-printer for machine learning research in IPython notebooks. |
| 59 | [dm_pix](https://github.com/google-deepmind/dm_pix) | 438 | active | 06/03/2 | PIX is an image processing library in JAX, for JAX. |
| 60 | [dm_robotics](https://github.com/google-deepmind/dm_robotics) | 411 | `archived` | 20/03/2 | Libraries, tools and tasks created and used at DeepMind Robotics. |
| 61 | [dm_env](https://github.com/google-deepmind/dm_env) | 400 | active | 23/12/2 | A Python interface for reinforcement learning environments |
| 62 | [spriteworld](https://github.com/google-deepmind/spriteworld) | 373 | active | 01/06/2 | Spriteworld: a flexible, configurable python-based reinforcement learning environment |
| 63 | [videoprism](https://github.com/google-deepmind/videoprism) | 372 | active | 14/05/2 | Official repository for "VideoPrism: A Foundational Visual Encoder for Video Understanding" (ICML 2024) |
| 64 | [regress-lm](https://github.com/google-deepmind/regress-lm) | 338 | active | 21/05/2 | Library for text-to-text regression, applicable to any input string representation and allows pretraining and fine-tuning over multiple regression tasks. |
| 65 | [launchpad](https://github.com/google-deepmind/launchpad) | 330 | `archived` | 22/08/2 | - |
| 66 | [kfac-jax](https://github.com/google-deepmind/kfac-jax) | 324 | active | 11/05/2 | Second Order Optimization and Curvature Estimation with K-FAC in JAX. |
| 67 | [enn](https://github.com/google-deepmind/enn) | 315 | active | 11/02/2 | - |
| 68 | [aloha_sim](https://github.com/google-deepmind/aloha_sim) | 313 | active | 13/11/2 | A collection of tabletop tasks in Mujoco |
| 69 | [leo](https://github.com/google-deepmind/leo) | 311 | active | 30/04/2 | Implementation of Meta-Learning with Latent Embedding Optimization |
| 70 | [nanodo](https://github.com/google-deepmind/nanodo) | 307 | active | 15/07/2 | - |
| 71 | [physics-IQ-benchmark](https://github.com/google-deepmind/physics-IQ-benchmark) | 293 | active | 05/05/2 | Benchmarking physical understanding in generative video models |
| 72 | [multi_object_datasets](https://github.com/google-deepmind/multi_object_datasets) | 288 | active | 03/03/2 | Multi-object image datasets with ground-truth segmentation masks and generative factors. |
| 73 | [gqn-datasets](https://github.com/google-deepmind/gqn-datasets) | 274 | active | 27/02/2 | Datasets used to train Generative Query Networks (GQNs) in the ‘Neural Scene Representation and Rendering’ paper. |
| 74 | [onetwo](https://github.com/google-deepmind/onetwo) | 268 | active | 15/04/2 | - |
| 75 | [grid-cells](https://github.com/google-deepmind/grid-cells) | 263 | active | 12/10/2 | Implementation of the supervised learning experiments in Vector-based navigation using grid-like representations in artificial agents, as published at https://www.nature.com/articles/s41586-018-0102-6 |
| 76 | [synjax](https://github.com/google-deepmind/synjax) | 250 | active | 20/02/2 | - |
| 77 | [loft](https://github.com/google-deepmind/loft) | 233 | active | 13/04/2 | LOFT: A 1 Million+ Token Long-Context Benchmark |
| 78 | [Temporal-3D-Pose-Kinetics](https://github.com/google-deepmind/Temporal-3D-Pose-Kinetics) | 227 | active | 30/08/2 | Exploiting temporal context for 3D human pose estimation in the wild: 3D poses for the Kinetics dataset |
| 79 | [dm_memorytasks](https://github.com/google-deepmind/dm_memorytasks) | 226 | active | 12/08/2 | A set of 13 diverse machine-learning tasks that require memory to solve. |
| 80 | [neural_networks_chomsky_hierarchy](https://github.com/google-deepmind/neural_networks_chomsky_hierarchy) | 218 | active | 12/04/2 | Neural Networks and the Chomsky Hierarchy |
| 81 | [jmp](https://github.com/google-deepmind/jmp) | 213 | active | 30/01/2 | JMP is a Mixed Precision library for JAX. |
| 82 | [magiclens](https://github.com/google-deepmind/magiclens) | 210 | active | 28/10/2 | [ICML'24 Oral] "MagicLens: Self-Supervised Image Retrieval with Open-Ended Instructions" |
| 83 | [dm_alchemy](https://github.com/google-deepmind/dm_alchemy) | 204 | `archived` | 05/04/2 | DeepMind Alchemy task environment: a meta-reinforcement learning benchmark |
| 84 | [predictingthepast](https://github.com/google-deepmind/predictingthepast) | 195 | active | 20/08/2 | - |
| 85 | [science-skills](https://github.com/google-deepmind/science-skills) | 185 | active | 21/05/2 | 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. |
| 86 | [language_modeling_is_compression](https://github.com/google-deepmind/language_modeling_is_compression) | 182 | active | 28/08/2 | - |
| 87 | [spectral_inference_networks](https://github.com/google-deepmind/spectral_inference_networks) | 171 | `archived` | 23/05/2 | Implementation of Spectral Inference Networks, ICLR 2019 |
| 88 | [jax_privacy](https://github.com/google-deepmind/jax_privacy) | 170 | active | 18/05/2 | Algorithms for Privacy-Preserving Machine Learning in JAX |
| 89 | [jaxline](https://github.com/google-deepmind/jaxline) | 166 | active | 13/12/2 | - |
| 90 | [interval-bound-propagation](https://github.com/google-deepmind/interval-bound-propagation) | 161 | active | 20/12/2 | This repository contains a simple implementation of Interval Bound Propagation (IBP) using TensorFlow: https://arxiv.org/abs/1810.12715 |
| 91 | [functa](https://github.com/google-deepmind/functa) | 161 | active | 30/07/2 | - |
| 92 | [md4](https://github.com/google-deepmind/md4) | 160 | active | 27/02/2 | Official Jax Implementation of MD4 Masked Diffusion Models |
| 93 | [jeo](https://github.com/google-deepmind/jeo) | 160 | active | 13/11/2 | Jeo: Jax model training lib for Earth Observation |
| 94 | [tensor_annotations](https://github.com/google-deepmind/tensor_annotations) | 159 | `archived` | 07/07/2 | Annotating tensor shapes using Python types |
| 95 | [language_to_reward_2023](https://github.com/google-deepmind/language_to_reward_2023) | 159 | active | 19/08/2 | - |
| 96 | [mishax](https://github.com/google-deepmind/mishax) | 157 | active | 16/02/2 | - |
| 97 | [jax_verify](https://github.com/google-deepmind/jax_verify) | 145 | `archived` | 18/08/2 | Neural network verification in JAX |
| 98 | [lamb](https://github.com/google-deepmind/lamb) | 138 | `archived` | 23/04/2 | LAnguage Modelling Benchmarks |
| 99 | [transformer_grammars](https://github.com/google-deepmind/transformer_grammars) | 137 | active | 20/05/2 | Transformer Grammars: Augmenting Transformer Language Models with Syntactic Inductive Biases at Scale, TACL (2022) |
| 100 | [conformal_training](https://github.com/google-deepmind/conformal_training) | 131 | `archived` | 16/08/2 | This repository contains a Jax implementation of conformal training corresponding to the ICLR'22 paper "learning optimal conformal classifiers". |
| 101 | [rgb_stacking](https://github.com/google-deepmind/rgb_stacking) | 129 | active | 30/07/2 | - |
| 102 | [tf2jax](https://github.com/google-deepmind/tf2jax) | 123 | active | 30/04/2 | - |
| 103 | [digraph_transformer](https://github.com/google-deepmind/digraph_transformer) | 122 | active | 13/09/2 | - |
| 104 | [bbeh](https://github.com/google-deepmind/bbeh) | 120 | active | 07/05/2 | - |
| 105 | [codoc](https://github.com/google-deepmind/codoc) | 120 | `archived` | 14/07/2 | - |
| 106 | [envlogger](https://github.com/google-deepmind/envlogger) | 116 | active | 19/04/2 | A tool for recording RL trajectories. |
| 107 | [geeflow](https://github.com/google-deepmind/geeflow) | 115 | active | 13/11/2 | GeeFlow - generate and process large-scale geospatial datasets with Google Earth Engine. |
| 108 | [einshape](https://github.com/google-deepmind/einshape) | 110 | active | 25/06/2 | - |
| 109 | [dm_env_rpc](https://github.com/google-deepmind/dm_env_rpc) | 109 | active | 20/02/2 | A networking protocol for agent-environment communication |
| 110 | [game_arena](https://github.com/google-deepmind/game_arena) | 106 | active | 02/02/2 | - |
| 111 | [dm_nevis](https://github.com/google-deepmind/dm_nevis) | 102 | active | 13/12/2 | NEVIS'22: Benchmarking the next generation of never-ending learners |
| 112 | [c3_neural_compression](https://github.com/google-deepmind/c3_neural_compression) | 99 | active | 13/04/2 | - |
| 113 | [habermas_machine](https://github.com/google-deepmind/habermas_machine) | 95 | active | 21/01/2 | - |
| 114 | [pushworld](https://github.com/google-deepmind/pushworld) | 94 | active | 05/05/2 | PushWorld: A benchmark for manipulation planning with tools and movable obstacles |
| 115 | [latent-multi-hop-reasoning](https://github.com/google-deepmind/latent-multi-hop-reasoning) | 91 | active | 18/03/2 | [ACL 2024] Do Large Language Models Latently Perform Multi-Hop Reasoning? |
| 116 | [emergent_in_context_learning](https://github.com/google-deepmind/emergent_in_context_learning) | 88 | active | 30/07/2 | - |
| 117 | [distribution_shift_framework](https://github.com/google-deepmind/distribution_shift_framework) | 87 | active | 19/05/2 | This repository contains the code of the distribution shift framework presented in A Fine-Grained Analysis on Distribution Shift (Wiles et al., 2022). |
| 118 | [dm_hard_eight](https://github.com/google-deepmind/dm_hard_eight) | 85 | active | 19/11/2 | - |
| 119 | [neural_networks_solomonoff_induction](https://github.com/google-deepmind/neural_networks_solomonoff_induction) | 84 | `archived` | 01/08/2 | Learning Universal Predictors |
| 120 | [randomized_positional_encodings](https://github.com/google-deepmind/randomized_positional_encodings) | 82 | active | 14/03/2 | Randomized Positional Encodings Boost Length Generalization of Transformers |
| 121 | [dks](https://github.com/google-deepmind/dks) | 79 | active | 01/07/2 | Multi-framework implementation of Deep Kernel Shaping and Tailored Activation Transformations, which are methods that modify neural network models (and their initializations) to make them easier to train. |
| 122 | [alignet](https://github.com/google-deepmind/alignet) | 76 | active | 01/12/2 | - |
| 123 | [dangerous-capability-evaluations](https://github.com/google-deepmind/dangerous-capability-evaluations) | 73 | active | 19/05/2 | - |
| 124 | [alphageometry2](https://github.com/google-deepmind/alphageometry2) | 73 | active | 07/01/2 | AlphaGeometry2 symbolic engine (DDAR) with examples |
| 125 | [proactive_t2i_agents](https://github.com/google-deepmind/proactive_t2i_agents) | 71 | active | 28/07/2 | Code release for the paper, "Proactive Agents for Text-to-Image Generation under Uncertainty" |
| 126 | [dmvr](https://github.com/google-deepmind/dmvr) | 69 | active | 03/11/2 | - |
| 127 | [dm_aux](https://github.com/google-deepmind/dm_aux) | 67 | active | 21/08/2 | - |
| 128 | [llms_can_learn_rules](https://github.com/google-deepmind/llms_can_learn_rules) | 63 | active | 06/12/2 | - |
| 129 | [detcon](https://github.com/google-deepmind/detcon) | 62 | `archived` | 29/10/2 | - |
| 130 | [pix2act](https://github.com/google-deepmind/pix2act) | 60 | active | 09/01/2 | - |
| 131 | [diplomacy](https://github.com/google-deepmind/diplomacy) | 60 | active | 22/04/2 | - |
| 132 | [calm](https://github.com/google-deepmind/calm) | 58 | active | 08/04/2 | - |
| 133 | [action_piece](https://github.com/google-deepmind/action_piece) | 58 | active | 08/04/2 | - |
| 134 | [natural-plan](https://github.com/google-deepmind/natural-plan) | 57 | active | 16/09/2 | - |
| 135 | [dm_fast_mapping](https://github.com/google-deepmind/dm_fast_mapping) | 54 | active | 28/10/2 | - |
| 136 | [alphatensor_quantum](https://github.com/google-deepmind/alphatensor_quantum) | 54 | active | 20/03/2 | - |
| 137 | [flows_for_atomic_solids](https://github.com/google-deepmind/flows_for_atomic_solids) | 53 | `archived` | 27/10/2 | - |
| 138 | [annealed_flow_transport](https://github.com/google-deepmind/annealed_flow_transport) | 53 | `archived` | 27/02/2 | - |
| 139 | [brave](https://github.com/google-deepmind/brave) | 51 | active | 17/03/2 | A JAX implementation of Broaden Your Views for Self-Supervised Video Learning, or BraVe for short. |
| 140 | [neural_lns](https://github.com/google-deepmind/neural_lns) | 51 | active | 03/02/2 | - |
| 141 | [streamingqa](https://github.com/google-deepmind/streamingqa) | 50 | active | 10/10/2 | - |
| 142 | [csuite](https://github.com/google-deepmind/csuite) | 47 | active | 24/09/2 | - |
| 143 | [dynamic-kanerva-machines](https://github.com/google-deepmind/dynamic-kanerva-machines) | 44 | active | 24/01/2 | This is a self-contained memory module for the Dynamic Kanerva Machine, as reported in the NIPS 2018 paper: Learning Attractor Dynamics for Generative Memory. |
| 144 | [disentangled_rnns](https://github.com/google-deepmind/disentangled_rnns) | 43 | active | 29/04/2 | - |
| 145 | [xarray_jax](https://github.com/google-deepmind/xarray_jax) | 42 | active | 07/04/2 | - |
| 146 | [ssl_hsic](https://github.com/google-deepmind/ssl_hsic) | 39 | active | 30/07/2 | - |
| 147 | [emergent_communication_at_scale](https://github.com/google-deepmind/emergent_communication_at_scale) | 39 | active | 30/07/2 | - |
| 148 | [questbench](https://github.com/google-deepmind/questbench) | 38 | active | 15/05/2 | - |
| 149 | [tell_me_why_explanations_rl](https://github.com/google-deepmind/tell_me_why_explanations_rl) | 37 | `archived` | 27/04/2 | - |
| 150 | [romo](https://github.com/google-deepmind/romo) | 37 | active | 25/09/2 | - |
| 151 | [fusion_surrogates](https://github.com/google-deepmind/fusion_surrogates) | 37 | active | 13/05/2 | A library of surrogate transport models for tokamak fusion. |
| 152 | [slim-dataset](https://github.com/google-deepmind/slim-dataset) | 36 | `archived` | 04/07/2 | Datasets used to train Spatial Language Integrating Model (SLIM) in the ‘Encoding Spatial Relations from Natural Language’ paper. |
| 153 | [dm_hamiltonian_dynamics_suite](https://github.com/google-deepmind/dm_hamiltonian_dynamics_suite) | 36 | active | 10/11/2 | - |
| 154 | [fancyflags](https://github.com/google-deepmind/fancyflags) | 35 | active | 04/07/2 | A Python library for defining structured command-line flags. |
| 155 | [eigengame](https://github.com/google-deepmind/eigengame) | 35 | `archived` | 15/05/2 | Open source code for EigenGame. |
| 156 | [spectral_ssm](https://github.com/google-deepmind/spectral_ssm) | 35 | active | 12/04/2 | - |
| 157 | [active_ops](https://github.com/google-deepmind/active_ops) | 33 | active | 30/07/2 | - |
| 158 | [hierarchical_perceiver](https://github.com/google-deepmind/hierarchical_perceiver) | 32 | active | 06/05/2 | - |
| 159 | [dmc_vision_benchmark](https://github.com/google-deepmind/dmc_vision_benchmark) | 32 | active | 21/06/2 | - |
| 160 | [proeval](https://github.com/google-deepmind/proeval) | 31 | active | 19/05/2 | GenAI evaluation framework, optimized for 100x lower cost 🚀. |
| 161 | [alta](https://github.com/google-deepmind/alta) | 31 | active | 22/09/2 | - |
| 162 | [qtqp](https://github.com/google-deepmind/qtqp) | 31 | active | 14/05/2 | - |
| 163 | [enn_acme](https://github.com/google-deepmind/enn_acme) | 30 | active | 25/08/2 | - |
| 164 | [slowfast_nfnets](https://github.com/google-deepmind/slowfast_nfnets) | 30 | `archived` | 22/06/2 | - |
| 165 | [inverse_design](https://github.com/google-deepmind/inverse_design) | 28 | active | 04/03/2 | - |
| 166 | [dm_construction](https://github.com/google-deepmind/dm_construction) | 28 | `archived` | 11/01/2 | - |
| 167 | [lm_act](https://github.com/google-deepmind/lm_act) | 28 | active | 21/05/2 | LMAct: A Benchmark for In-Context Imitation Learning with Long Multimodal Demonstrations |
| 168 | [zipfian_environments](https://github.com/google-deepmind/zipfian_environments) | 28 | active | 28/07/2 | - |
| 169 | [image_obfuscation_benchmark](https://github.com/google-deepmind/image_obfuscation_benchmark) | 27 | active | 20/05/2 | - |
| 170 | [trecvit](https://github.com/google-deepmind/trecvit) | 26 | active | 10/01/2 | - |
| 171 | [simulation_streams](https://github.com/google-deepmind/simulation_streams) | 25 | active | 13/03/2 | Simulation Streams is a programming paradigm designed to efficiently control and leverage Large Language Models (LLMs) for complex, dynamic simulations and agentic workflows. |
| 172 | [sam_edge](https://github.com/google-deepmind/sam_edge) | 24 | active | 20/05/2 | - |
| 173 | [wasserstein_fairness](https://github.com/google-deepmind/wasserstein_fairness) | 23 | active | 26/01/2 | An implementation of Wasserstein Fair Classification, a conference paper submitted to UAI 2019. |
| 174 | [neural_assets](https://github.com/google-deepmind/neural_assets) | 22 | active | 08/05/2 | - |
| 175 | [affordances_option_models](https://github.com/google-deepmind/affordances_option_models) | 22 | active | 08/11/2 | - |
| 176 | [iris](https://github.com/google-deepmind/iris) | 21 | active | 04/06/2 | - |
| 177 | [protex](https://github.com/google-deepmind/protex) | 20 | active | 19/09/2 | - |
| 178 | [egg](https://github.com/google-deepmind/egg) | 20 | active | 15/04/2 | - |
| 179 | [agent_debugger](https://github.com/google-deepmind/agent_debugger) | 20 | active | 27/06/2 | Causal Analysis of Agent Behavior for AI Safety |
| 180 | [deep-verify](https://github.com/google-deepmind/deep-verify) | 19 | active | 11/11/2 | - |
| 181 | [nao_top10](https://github.com/google-deepmind/nao_top10) | 19 | active | 01/03/2 | - |
| 182 | [unlearning_evaluation](https://github.com/google-deepmind/unlearning_evaluation) | 18 | active | 30/05/2 | - |
| 183 | [threednel](https://github.com/google-deepmind/threednel) | 18 | active | 10/11/2 | - |
| 184 | [geomatch](https://github.com/google-deepmind/geomatch) | 18 | active | 11/12/2 | - |
| 185 | [serial_depth](https://github.com/google-deepmind/serial_depth) | 18 | active | 10/03/2 | - |
| 186 | [nonstationary_mbml](https://github.com/google-deepmind/nonstationary_mbml) | 17 | active | 14/03/2 | Memory-Based Meta-Learning on Non-Stationary Distributions |
| 187 | [codesembench](https://github.com/google-deepmind/codesembench) | 16 | `archived` | 22/03/2 | - |
| 188 | [icml2024-roundtrip-correctness](https://github.com/google-deepmind/icml2024-roundtrip-correctness) | 16 | `archived` | 18/06/2 | - |
| 189 | [scivid](https://github.com/google-deepmind/scivid) | 16 | active | 02/03/2 | - |
| 190 | [eval_hub](https://github.com/google-deepmind/eval_hub) | 16 | active | 19/02/2 | Gemini evaluations |
| 191 | [ccbo](https://github.com/google-deepmind/ccbo) | 15 | active | 18/06/2 | This repo contains the code associated to the paper:  "Constrained Causal Bayesian Optimization" by Aglietti Virginia, Alan Malek, Ira Ktena, and Silvia Chiappa. International Conference on Machine Learning. PMLR, 2023. |
| 192 | [exedec](https://github.com/google-deepmind/exedec) | 14 | active | 09/05/2 | - |
| 193 | [recoglab](https://github.com/google-deepmind/recoglab) | 13 | active | 19/05/2 | - |
| 194 | [tanq](https://github.com/google-deepmind/tanq) | 11 | active | 20/06/2 | - |
| 195 | [autonumerics_zero](https://github.com/google-deepmind/autonumerics_zero) | 11 | active | 13/04/2 | - |
| 196 | [anthro-benchmark](https://github.com/google-deepmind/anthro-benchmark) | 11 | active | 17/11/2 | - |
| 197 | [strategicwm](https://github.com/google-deepmind/strategicwm) | 11 | active | 19/05/2 | - |
| 198 | [constrained_optidice](https://github.com/google-deepmind/constrained_optidice) | 10 | active | 09/09/2 | - |
| 199 | [semppl](https://github.com/google-deepmind/semppl) | 10 | `archived` | 18/04/2 | - |
| 200 | [max_product_noisy_or](https://github.com/google-deepmind/max_product_noisy_or) | 10 | active | 20/05/2 | - |
| 201 | [amld_workshop_natural_interactions_with_llms](https://github.com/google-deepmind/amld_workshop_natural_interactions_with_llms) | 10 | active | 17/04/2 | - |
| 202 | [implicit_diffusion](https://github.com/google-deepmind/implicit_diffusion) | 10 | active | 03/03/2 | - |
| 203 | [thunnini](https://github.com/google-deepmind/thunnini) | 10 | active | 17/10/2 | Experimentation library for comparing fine-tuners for neural sequential predictors. |
| 204 | [nfg_transformer](https://github.com/google-deepmind/nfg_transformer) | 9 | active | 14/02/2 | - |
| 205 | [cube](https://github.com/google-deepmind/cube) | 8 | active | 20/01/2 | Code for CUBE extraction and cultural diversity metric introduced in "Beyond Aesthetics: Cultural Competence in Text-to-Image Models" |
| 206 | [leaps-and-bounds](https://github.com/google-deepmind/leaps-and-bounds) | 8 | active | 26/01/2 | This is an implementation of LeapsAndBounds and Structured Procrastination for approximately optimal algorithm configuration. |
| 207 | [offpolicy_selection_eslb](https://github.com/google-deepmind/offpolicy_selection_eslb) | 8 | active | 17/03/2 | - |
| 208 | [rem](https://github.com/google-deepmind/rem) | 8 | active | 04/02/2 | - |
| 209 | [gnn_single_rigids](https://github.com/google-deepmind/gnn_single_rigids) | 7 | active | 04/03/2 | - |
| 210 | [dm_c19_modelling](https://github.com/google-deepmind/dm_c19_modelling) | 7 | `archived` | 20/07/2 | - |
| 211 | [meqpy](https://github.com/google-deepmind/meqpy) | 7 | active | 24/03/2 | - |
| 212 | [corr_faith](https://github.com/google-deepmind/corr_faith) | 6 | active | 29/04/2 | - |
| 213 | [atomic_concept_edits](https://github.com/google-deepmind/atomic_concept_edits) | 6 | active | 04/04/2 | - |
| 214 | [mona](https://github.com/google-deepmind/mona) | 6 | active | 14/04/2 | - |
| 215 | [platonic_rep_video](https://github.com/google-deepmind/platonic_rep_video) | 6 | active | 22/04/2 | - |
| 216 | [polarix](https://github.com/google-deepmind/polarix) | 5 | active | 02/02/2 | - |
| 217 | [batch-isolation-checker](https://github.com/google-deepmind/batch-isolation-checker) | 5 | active | 20/03/2 | - |
| 218 | [tsuite](https://github.com/google-deepmind/tsuite) | 5 | active | 24/09/2 | - |
| 219 | [mir_uai25](https://github.com/google-deepmind/mir_uai25) | 5 | active | 04/11/2 | - |
| 220 | [relicv2](https://github.com/google-deepmind/relicv2) | 5 | active | 16/12/2 | - |
| 221 | [additive_cbug](https://github.com/google-deepmind/additive_cbug) | 5 | `archived` | 06/06/2 | - |
| 222 | [abcei_mab](https://github.com/google-deepmind/abcei_mab) | 5 | active | 12/05/2 | - |
| 223 | [local_linearity_regularizer](https://github.com/google-deepmind/local_linearity_regularizer) | 4 | active | 30/07/2 | - |
| 224 | [mammut](https://github.com/google-deepmind/mammut) | 4 | active | 18/06/2 | - |
| 225 | [privately-counting-distinct-elements](https://github.com/google-deepmind/privately-counting-distinct-elements) | 3 | active | 18/04/2 | Code accompanying "Counting Distinct Elements Under Person-Level Differential Privacy" |
| 226 | [sketch_dqn](https://github.com/google-deepmind/sketch_dqn) | 3 | active | 13/04/2 | - |
| 227 | [utm](https://github.com/google-deepmind/utm) | 3 | active | 03/12/2 | - |
| 228 | [what_type_of_inference_is_planning](https://github.com/google-deepmind/what_type_of_inference_is_planning) | 3 | active | 07/12/2 | - |
| 229 | [beneath-the-surface](https://github.com/google-deepmind/beneath-the-surface) | 3 | active | 15/04/2 | - |
| 230 | [unpuzzles_and_simple_reasoning](https://github.com/google-deepmind/unpuzzles_and_simple_reasoning) | 3 | active | 09/09/2 | - |
| 231 | [llp_bp](https://github.com/google-deepmind/llp_bp) | 2 | active | 01/06/2 | - |
| 232 | [agg_data_uai25](https://github.com/google-deepmind/agg_data_uai25) | 2 | active | 30/06/2 | - |
| 233 | [ddgc](https://github.com/google-deepmind/ddgc) | 2 | active | 24/04/2 | - |
| 234 | [wtos_agglabels_uai25](https://github.com/google-deepmind/wtos_agglabels_uai25) | 2 | active | 26/06/2 | - |
| 235 | [covariate_shifted_llp](https://github.com/google-deepmind/covariate_shifted_llp) | 2 | active | 12/06/2 | Code accompanying UAI'25 paper titled "Learning from Label Proportions and Covariate-shifted Instances" |
| 236 | [fractal_acl25](https://github.com/google-deepmind/fractal_acl25) | 2 | active | 26/06/2 | - |