Basics
Most University licences are confidential, and so they should not be uploaded to cloud hosted LLMs such as Claude since that act effectively makes them public and breaks confidentiality.
There are two ways to get round this. First is to spend time redacting the University licence to remove all information about the University and your company. This is time consuming and error prone, and it’s absolutely not recommended.
The alternative is to download and run a local SLM and then block access to the internet either by simply disconnecting all network connection (e.g. run in Airplane mode) or use a Firewall to block the SLM making outbound connections.
There are various local SLMs such as Ollama which can be used. The examples below use LM Studio.
Download and Installation
LM Studio is available at https://lmstudio.ai/download
Click the Download LM Studio button (not Bionic) and follow the installation instructions.
For those Arm fans out there, there is also an Arm (Snapdragon) variant, and the models have also been tested on those.
Models
Models are updated regularly and for new ones look in Model Search. The prompt examples below use the qwen/qwen3.5-9b model, which is a less than 10 Gbyte sized model that should run within most desktop environments.
If you have a desktop with more memory, then feel free to choose a larger sized model, but note that we have tested the qwen/qwen3.5-9b model against multiple different licenses, and multiple different models and it’s hard to beat.
Note that since the models are many Gbytes large, then it may make sense to use a drive other than C: such as D:. In this case, go to the “My Models” tab (folder icon), click Settings and change the Models Directory path to a folder that you have created on your D: drive.
Settings
Default settings are usually fine, but changing the Default Context Length setting in Model Defaults provides a larger context, which is useful for the Qwen model and reading legal documents.
Helping in negotiations
The Prompt file incorporates the key negotiation points in the Framework.
Upload your chosen Licence and the Prompt into the “Send a message to the model …” window either by dragging and dropping them, or using the + entry in the window, and then just hit return. You should then see something very similar to the message below. The key requirement is that the model chooses “context injection strategy: ‘inject-full-context’”. That means it will fully read the Licence and the Prompt.
Note that while running locally is a secure process, it is not fast, and will take 20-30 mins, and you will have to close down most other applications to give enough memory to run the model. So make sure you have changed the power plan so your machine doesn’t go to sleep, and get a cup of coffee.
Once it runs, you will have a complete set of differences from the SDL and best practice, and more importantly, a set of recommendations, justifications and specific actions to take to your TTO to make the negotiations more fruitful and hopefully quicker.
You can then ask further clarification questions of the model such as “If we do a funding raise of £500K in 6 months and have zero revenue, then how do the licences compare in terms of cash outlay by the Company”, etc. This helps you work through different scenarios.