SpicyChat AI Models: Free and Paid Options, Settings and Context

Choose for the kind of conversation you want

A model that suits brief dialogue may not be your preference for a slow, descriptive scene. Test the same short premise with the models available to your account. Keep your greeting and persona unchanged so that you can tell which differences come from the model.

Look at whether the response answers the last message, respects the speaker's point of view, and introduces a useful next step. More words are not automatically a better continuation.

Examples of model access by plan

Higher tiers include the lower-tier choices. This is a selection of our models, not a promise that a retired model will remain available. The Change Model menu shows the choices currently enabled for your account.

Size, context, and access describe different things

A parameter count describes model size. Context describes how much information can be used for a reply. A subscription determines which model and feature choices your account can access. One large number does not tell you all three.

The effective context cannot exceed either the model's capacity or your tier's allowance. Choosing a model with a larger technical window does not by itself raise a Free account's context limit.

Change one variable at a time

Open the three-dot chat menu, choose Generation Settings, then find Inference Model and select Change Model. After switching, try a few ordinary exchanges before changing temperature, reply length, and the character definition as well. Changing all of them together makes it difficult to identify the cause of an improvement or problem.

A model can still make factual mistakes, repeat language, or lose an older detail. For a story-continuity problem, review the relevant memory or instruction instead of treating a model change as a guaranteed fix.