SpicyChat Model Choices: Plan Access, Context and Switching

Compare models with the same short scene

Keep the greeting and persona fixed while trying the models your account offers. That makes it easier to compare rapid back-and-forth dialogue with more descriptive storytelling. Your preference may depend on the conversation rather than on one model suiting every style.

Check whether the reply addresses your last message, keeps the speakers' viewpoints clear and gives the scene a useful next step. Length alone is not a measure of how well the continuation fits.

See examples at each entry tier

Higher tiers also include the choices from the tiers below them. These examples are a selection, not a permanent or exhaustive catalogue. Use Change Model to see what is currently enabled for your account; an older list does not guarantee continued access to a retired model.

Read model size and context as separate limits

Parameter counts describe a model's size, not its context allowance or the subscription needed to select it. Context concerns the information available for a reply, while your plan controls access to models and features. Those are different questions even when each is described with a large number.

Your effective context is constrained by both the model's capacity and your tier's allowance. Selecting a model with a technically larger window does not increase a Free account's entitlement on its own.

Switch first, then assess other settings

In the three-dot chat menu, open Generation Settings, find Inference Model and choose Change Model. Try a few typical exchanges after switching. If you also change temperature, reply length and the character definition immediately, it becomes hard to tell which adjustment affected the result.

A different model can still repeat itself, miss an older detail or make a factual error. For continuity problems, check the relevant instructions and memories as well. Switching models is an option to test, not a guaranteed repair.