The models below run on your own device. This site provides the interface that downloads and runs them, so the terms each model carries are passed on to you here. By using a model on this site you agree to the terms that apply to it.
Gemma (Google)
Gemma is provided under and subject to the Gemma Terms of Use found at ai.google.dev/gemma/terms.
- Gemma 3 1B Instruct and Gemma 2 2B Instruct, quantised to 4-bit by the MLC project.
- Full terms: ai.google.dev/gemma/terms
- Use of Gemma is subject to the restrictions in the Gemma Prohibited Use Policy, which applies to you when you use Gemma through this site. You must not use Gemma for any purpose that policy forbids.
- Gemma is a trademark of Google LLC. This site is not affiliated with, endorsed by, or sponsored by Google.
Llama 3.2 (Meta)
Built with Llama.
- Llama 3.2 1B Instruct, quantised to 4-bit by the MLC project.
- Licensed under the Llama 3.2 Community License, Copyright © Meta Platforms, Inc. All Rights Reserved. Use is also subject to the Llama 3.2 Acceptable Use Policy.
- Llama is a trademark of Meta Platforms, Inc. This site is not affiliated with, endorsed by, or sponsored by Meta.
SmolLM2 (Hugging Face)
- SmolLM2 360M Instruct, quantised to 4-bit by the MLC project.
- Copyright © Hugging Face. Licensed under the Apache License, Version 2.0.
WebLLM / MLC (this site serves a copy)
The engine that runs the models in your browser is WebLLM by the MLC project. A copy of it is served from this domain at /web-llm.js, so a copy of its licence is provided here as required.
- Copyright © MLC-AI / WebLLM contributors.
- Licensed under the Apache License, Version 2.0 — a full copy is served at /licences/apache-2.0.txt.
- Model weights are downloaded by your browser from the Hugging Face CDN and are not redistributed by this site.
This site
The interface itself is the work of the operator of ai.koas.uk. The models are the work of their respective creators, named above, and are not the operator's work.