Open versus closed models, explained fairly
Understand open weights, open source and closed APIs, and how to choose between them
A taste of a lesson
If a model is open, can I use it for anything I want in my business?
Not automatically. 'Open' often means open weights: you can download and run it, but the licence decides what you may do. Some licences are permissive and allow commercial use and modification. Others add conditions: usage policies, limits above a certain number of users, attribution, or bans on using outputs to train competing models. So the first step is always reading the licence, ideally with your legal team for anything commercial. Exercise: list the three licence terms you would check first for your business.
Written by the teacher as an example. In your lesson the tutor answers your own questions, and like any AI it can be wrong.
What you will be able to do
- Define closed, open weights and open source AI precisely
- Read a model licence summary for commercial use and restrictions
- Compare open and closed options on capability, cost, privacy, control and support
- Present the main arguments in the debate over releasing model weights
Lesson plan
- 1 What 'open' really means Pin down the terms closed, open weights and open source. Start
- 2 Reading a licence Find the terms that matter in a model licence. Start
- 3 Trade offs for real decisions Compare open and closed options on practical criteria. Start
- 4 The release debate Weigh the arguments for and against releasing powerful model weights. Start
- 5 Choosing for a situation Apply a decision checklist to real scenarios. Start
Try asking
About this tutor
For anyone weighing open and closed AI models: learners, managers, developers and policy minded readers. You learn what 'open' actually means (open weights, open source code, open data, and licence terms that differ widely), how closed models are accessed through apps and APIs, and the trade offs in capability, cost, privacy, control, customisation, support and safety. You also look at the debate over whether releasing powerful model weights is good or risky, presented fairly from several sides. You finish able to read a model licence summary and choose the right kind of model for a situation.
Reviews
4.3
3 ratingsSample
- Pierre L.Sample
Good, balanced. I wanted more detail on hardware costs for running open models, but that is covered elsewhere I think.
- Jonas B.Sample
The distinction between open weights and open source cleared up months of confusion in our team. Licence reading practice was very practical.
- Adaeze O.Sample
Fair on the release debate, which is rare. The scenarios lesson helped me make a case for a hybrid approach at work.
About the teacher
I explain the kinds of AI models, what they cost to run and how to run one yourself
9 tutors 328 lessons taught Sample
I teach the practical side of modern models: reasoning models, multimodal models, open and closed weights, running a model on your own computer, and the money, energy and hardware behind every answer. I like starting with something you can see or measure, such as the memory a model needs or the number of tokens a task uses, then explaining the...
See Benedict's profile and tutorsMore like this
Other tutors on the same or nearby topics.