A scalable asynchronous reinforcement learning implementation with in-flight weight updates.

A high-throughput and memory-efficient inference and serving engine for LLMs

A Python library for fast and easy access to genomic resources such as sequence, data tracks, and annotations

Implement a ChatGPT-like LLM in PyTorch from scratch, step by step

Open-source framework for the research and development of foundation models.

⚡ TabPFN: Foundation Model for Tabular Data ⚡

Library for efficient training and application of Machine Learning Interatomic Potentials (MLIP)

[NeurIPS 2025 spotlight] Efficient factorized variant of the IPA module.

RNAGym is an extensive benchmark suite and resource for RNA fitness and structure prediction

GenAI Agent Framework, the Pydantic way

Refine.bio harmonizes petabytes of publicly available biological data into ready-to-use datasets for cancer researchers and AI/ML scientists.

A benchmark for comprehensive evaluation on protein structure tokenization methods

Inference code for scalable emulation of protein equilibrium ensembles with generative deep learning

Results and data from the pilot round of the Protein Engineering Tournament

Minimalistic 4D-parallelism distributed training framework for education purpose

Nature Methods: RNA foundation model (together with RhoFold)

🧬 Augmenting zero-shot mutant prediction by retrieval-based logits fusion. (ISMB/ECCB 2025)

Sire Molecular Simulations Framework

Roo Code gives you a whole dev team of AI agents in your code editor.

Remote MCP Servers

A high-performance distributed file system designed to address the challenges of AI training and inference workloads.

Genomic Pre-trained Network

🐙 A curated database of completed assemblies with taxonomy IDs

Easily fine-tune, evaluate and deploy gpt-oss, Qwen3, DeepSeek-R1, or any open source LLM / VLM!

Explorations into whether a transformer with RL can direct a genetic algorithm to converge faster

Fully open reproduction of DeepSeek-R1

MassSpecGym: A benchmark for the discovery and identification of molecules (NeurIPS 2024 Spotlight)

Official repo of the modular BioExcel version of HADDOCK

Sky-T1: Train your own O1 preview model within $450