Local models in use by Shade

Window sizes vary by model, so Shade needs to carefully craft the context it sends to each model

List of local models that are currently used by Shade, each with various parameter sizes and context window lengths. The gemma3:1b model, for example, has a context window of 32K tokens, while another 1B param model, llama3.2:1b, has a context window of 128K tokens. A model's parameter size isn't always a good indication of the size of its context window.
Get the dataCreated with Datawrapper