A thoughtful AI collaborator for writing, analysis, coding, and long-form work.
Full evaluationGlossary · Updated Sep 2, 2026
Open-weight model
A model whose trained parameters are published so anyone can download, run, and often fine-tune it, subject to a licence.
Definition
Open-weight models, sometimes loosely called open source, let organisations run AI on their own infrastructure, inspect behaviour, and customise through fine-tuning. Licences vary from permissive to restricted commercial use, and weights alone do not include training data. Open models underpin self-hosted assistants, image generators, and automation platforms.
Why it matters when choosing a tool
Open weights give control over privacy, cost, and customisation, at the price of operating the models yourself.
Where you will meet it
AI Coding & Development, AI Image & Design, AI Assistants
Related terms
Self-hosting · Fine-tuning · LoRA (low-rank adaptation) · Data residency and privacy
Tools where this matters
Reviewed products in the related categories
An AI-native code editor for repository-aware agents, edits, review, and automation.
Full evaluation
Anthropic's agentic coding tool that works in the terminal, IDE, desktop app, and browser to plan and execute multi-step changes.
Full evaluation
AI creation inside a broad collaborative design and content production suite.
Full evaluation
An AI-first workspace for creating presentations, documents, websites, and visual stories.
Full evaluation
A generative video and creative production platform for controllable AI filmmaking.
Full evaluation