facebookresearch/neuroai

Python suite for neuroscience research across all modalities.

What it solves

Neuro AI provides a suite of tools to simplify the process of building, training, and evaluating models that bridge neuroscience and artificial intelligence. It addresses the complexity of handling specialized neuro-AI data, scaling training, and maintaining consistent benchmarking across different models.

How it works

The project is organized into four specialized Python packages:

  • NeuralSet: A high-performance data loader designed specifically for Neuro AI datasets.
  • NeuralFetch: A tool for fetching curated Neuro AI datasets.
  • NeuralTrain: A framework for training Neuro AI models at scale.
  • NeuralBench: A unified benchmarking system to evaluate Neuro AI models.

Who it’s for

Researchers and developers working at the intersection of neuroscience and artificial intelligence who need efficient data pipelines and scalable training infrastructure.

Highlights

  • Modular ecosystem: Separate packages for data loading, data acquisition, and training.
  • Scalable training: Dedicated tooling via NeuralTrain for large-scale model development.
  • Unified benchmarking: Standardized evaluation through NeuralBench.
  • High-performance loading: Optimized data handling via NeuralSet.

Related

  • Project
    Datayoo/HuggingFists

    A low-code data flow tool that enables users to visually build and schedule workflows using LLMs and Hugging Face models without extensive coding.

  • Project
    sopaco/cortex-mem

    A Rust-based long-term memory framework for AI agents that uses a three-tier hierarchical architecture and a virtual filesystem to provide persistent, context-aware interactions.

  • Project
    finna/Finn-loop

    A set of Claude Code skills that automate the pipeline from idea to PR using Linear and GitHub, while keeping humans as the final merge authority.

  • Project
    meilisearch/arroy

    A Rust library for approximate nearest neighbor search that uses LMDB for low-memory, memory-mapped vector indexing and retrieval.