Search Results

Welcome to the Vector Institute Bookstack Wiki! The Vector Bookstack is a repository built and maintained by the AI Engineering team to share knowledge and documentation both internally and externally! See Contributing for guidelines on the Wiki structure bef...

Guidelines for contributing documentation. For tips on editing such as custom callouts see Editing Tips. Bookstack Basics: Documentation in Bookstack is organized into Books. Each Book contains Pages which contain documentation. Pages within a book can be gro...

The Vector Inference Platform is a service provided by the AI Engineering team at Vector Institute. It hosts large, state-of-the-art open-source language models that anyone in the Vector community can use freely and easily.Unlike previous efforts to provide in...

The Vector Inference Platform is available to all Vector Institute community members. For an up-to-date list of available models and their specifications, visit inference.vectorinstitute.ai.Getting an API KeyAccess is managed via invite links. To get started:C...

This page outlines some tips and guidelines to use when editing pages. Choosing an Editor The default editor for this Bookstack wiki has been set to Markdown. If you prefer a more traditional WYSIWYG editor (What You See Is What You Get) similar to microsoft w...

Starting in early 2026, the Scientific Computing is implementing updates to the Bon Echo environment that will affect how users access the cluster and run workflows there. Notable changes include: The original login nodes (ie. v.vectorinstitute.ai and vlogin....

Prerequisites Python 3.10 or newer uv for dependency management Setting Up Development Environment Clone the repository: git clone https://github.com/VectorInstitute/vector-inference.git cd vector-inference Install development dependencies: uv sync --al...

Development Workflow Branch off of main, name your branch with the following convention: f/$FEATURE_NAME, b/$BUG_FIX or r/RELEASE_TAG. Make new changes in your branch, and add/modify relevant unit tests in tests. Please follow the styling guide in the next se...

Release PR It is recommended to create a dedicated branch and PR for a new release: Update version in pyproject.toml. Run uv lock --upgrade to update the lock file. Pull new images on the cluster if inference engine version is updated, more details in next se...

Cached Configs and Apptainer Images In the /model-weights folder, the following files exists: environment.yaml: Cached environment config models.yaml: OLD cached model config. This is from pre-v0.8.0, it is no longer being updated, cached model config path wa...