Stars
Structure-based fragment identification in latent space
Official implementation of MatterGen -- a generative model for inorganic materials design across the periodic table that can be fine-tuned to steer the generation towards a wide range of property c…
cuEquivariance is a math library that is a collective of low-level primitives and tensor ops to accelerate widely-used models, like DiffDock, MACE, Allegro and NEQUIP, based on equivariant neural n…
Official repository for the ProteinGym benchmarks
BioDiscoveryAgent is an LLM-based AI agent for closed-loop design of genetic perturbation experiments
BioNeMo Recipes: For building and adapting AI models in drug discovery at scale
MACE - Fast and accurate machine learning interatomic potentials with higher order equivariant message passing.
MaSIF- Molecular surface interaction fingerprints. Geometric deep learning to decipher patterns in molecular surfaces.
A benchmark for 3D biomolecular structure prediction models
tiktoken is a fast BPE tokeniser for use with OpenAI's models.
Discovering Interpretable Features in Protein Language Models via Sparse Autoencoders
Official repository for the Boltz biomolecular interaction models
LatentDE: Latent-based Directed Evolution for Protein Sequence Design
AlphaFold 3 inference pipeline.
Toward High-Accuracy Open-Source Biomolecular Structure Prediction.
[ICML 25] We train and evaluate SAEs to identify interpretable features in pLMs and show their potential for scientific discovery.
A list of interesting genome browser and genome visualization programs
RWKV (pronounced RwaKuv) is an RNN with great LLM performance, which can also be directly trained like a GPT transformer (parallelizable). We are at RWKV-7 "Goose". So it's combining the best of RN…
Representation Engineering: A Top-Down Approach to AI Transparency
Repository for StripedHyena, a state-of-the-art beyond Transformer architecture
Training Sparse Autoencoders on Language Models
Create feature-centric and prompt-centric visualizations for sparse autoencoders (like those from Anthropic's published research).
A high-throughput and memory-efficient inference and serving engine for LLMs
Generative modeling of molecular dynamics trajectories
DSPy: The framework for programming—not prompting—language models
A virtual environment for developing and evaluating automated scientific discovery agents.