Skip to content
@VectorInstitute

Vector Institute

README.md

Welcome to Vector Institute's github!

At the Vector Institute, we are dedicated to advancing artificial intelligence (AI) research and translating cutting-edge innovations into real-world solutions. Our open-source projects reflect our commitment to collaboration, transparency, and the responsible deployment of AI technologies. Here, you’ll find tools, frameworks, and resources that empower organizations and researchers to harness the full potential of AI.

What We Build

Our open-source contributions aim to bridge the gap between theoretical research and practical application of AI. We build tools, MVPs, reference implementations and educational resources that empower researchers, developers, and organizations to innovate and solve real-world problems with AI. Our projects span various domains and are designed to be accessible, adaptable, and impactful.

Here are a few key projects that exemplify our mission:

Vector Inference

VectorInstitute/vector-inference is a project focused on providing efficient and streamlined LLM inference on Slurm clusters. It provides a simple command-line interface and examples for deploying LLMs using vLLM.

AI Pocket Reference

  • The VectorInstitute/ai-pocket-reference serves as a concise and accessible guide to essential AI concepts, algorithms, and techniques. It's designed to be a quick reference for practitioners and students, covering a wide range of topics ranging from basics to advanced SOTA approaches.

Reference Implementation Catalog

  • The VectorInstitute/reference-implementation-catalog offers a curated collection of reference implementations for various AI algorithms and models. This catalog provides practical, working examples that demonstrate how to implement these techniques in code. It's an invaluable resource for those looking to understand and apply AI algorithms to their problems and datasets.

Why Open Source?

We believe that open-source AI tools are essential for fostering innovation and addressing global challenges. By sharing our work, we aim to:

  • Accelerate AI Adoption: Provide accessible tools that lower barriers to entry for organizations and researchers.
  • Drive Collaboration: Encourage partnerships across academia, industry, and government to solve complex problems.
  • Promote Responsible AI: Ensure that AI technologies are developed and deployed ethically, with transparency and accountability.

Get Involved

We welcome contributions from the global AI community. Whether you’re a researcher, developer, or industry professional, there are many ways to engage with our projects:

  • Explore Our Repositories: Dive into our code and documentation to understand how our tools work.
  • Contribute: Submit issues, suggest features, or submit pull requests to help improve our projects.
  • Collaborate: Partner with us on new initiatives or adapt our tools to your needs.

Connect With Us

Stay updated on our latest projects and announcements:

Together, we can unlock the transformative potential of AI and drive meaningful impact across industries and society.

Pinned Loading

  1. vectorlm Public

    LLM finetuning in resource-constrained environments.

    Python 47 9

Repositories

Showing 10 of 131 repositories
  • kg-rag Public

    This project implements a comprehensive framework for Knowledge Graph Retrieval Augmented Generation (KG-RAG). It focuses on financial data from SEC 10-Q filings and explores how knowledge graphs can improve information retrieval and question answering compared to baseline approaches.

    Python 5 1 0 0 Updated May 1, 2025
  • FL4Health Public

    A flexible, modular, and easy to use library to facilitate federated learning research and development in healthcare settings

    Python 39 Apache-2.0 12 0 4 Updated May 1, 2025
  • fed-rag Public

    A framework for federated fine-tuning of retrieval-augmented generation (RAG) systems.

    Python 10 Apache-2.0 1 38 (5 issues need help) 2 Updated May 1, 2025
  • vector-inference Public

    Efficient LLM inference on Slurm clusters using vLLM.

    Python 58 MIT 10 3 5 Updated May 1, 2025
  • Vector-Jobboard Public

    Working with Jobboard API to perform actions needed by Vector Institute

    Python 0 0 0 1 Updated May 1, 2025
  • shared-encoder Public

    Codebase for the paper titled 'A Shared Encoder Approach to Multimodal Representation Learning'

    Python 8 MIT 1 0 2 Updated May 1, 2025
  • FLorist Public
    CSS 8 Apache-2.0 0 0 0 Updated Apr 30, 2025
  • health-rec Public

    MVP of a recommendation system for health and community services

    Python 6 0 5 0 Updated Apr 30, 2025
  • cyclops Public

    A toolkit for evaluating and monitoring AI models in clinical settings

    Python 83 13 0 8 Updated Apr 29, 2025
  • AtomGen Public

    Library for handling atomistic graph datasets focusing on transformer-based implementations, with utilities for training various models, experimenting with different pre-training tasks, and a suite of pre-trained models with huggingface integrations

    Python 5 Apache-2.0 0 0 4 Updated Apr 29, 2025
View all repositories