Auto Model Routing Explained — How SoulrisAI Picks the Right AI for You
What "Auto" actually means
By default, SoulrisAI doesn't send every message to the same AI model. Instead, Auto routing looks at your message and decides which model is actually suited to answer it — a quick factual question doesn't need the same model as a complex technical or creative request. This happens automatically, in the background, before you see a response.
Why route at all instead of using one model for everything?
Different AI models are genuinely better at different things, and they cost different amounts to run. Always using the strongest available model for every single message would make chat far more expensive than it needs to be for simple questions, while always using the cheapest model would give worse answers on anything genuinely hard. Routing is how SoulrisAI gives you a good answer without you having to think about which model that requires.
The tiers behind the scenes
SoulrisAI organizes models into tiers — free, basic, intelligent, expert, and super — roughly from lightest/cheapest to most capable/expensive. Auto routing selects among these based on your message, so a simple "what's the capital of Nigeria?" might be answered by a lighter model, while "review this contract clause for risk" is more likely to route to a stronger one. See understanding chat tiers for what each tier is actually good for.
Do I need to understand this to use chat well?
No — this is the point of Auto routing. It exists so you can just ask what you actually need to ask, in plain language, without first deciding "is this a basic question or an expert one?" The routing logic does that judgment call for you, message by message.
When you might want more control
Auto routing is the right default for most people, most of the time. But if you consistently want a specific model regardless of how the question looks — maybe you've found one model's writing style suits you, or you want to compare two models on the same prompt — you can override Auto and pick manually. See how to manually choose a specific model for how.
Does routing affect cost?
Yes, indirectly. Stronger-tier models generally cost more per response than lighter ones, and Auto routing's whole purpose is to only use a stronger (more expensive) model when your message actually seems to need it — so letting Auto decide is usually the more cost-efficient choice, not just the simpler one. Manually forcing every message to the top tier will cost more without necessarily improving the answer to a simple question.
A practical way to think about it
If you're not sure whether something counts as a "hard" question, don't worry about it — just ask it normally. Auto routing exists precisely so that judgment isn't yours to make. Spend your attention on writing a clear, specific message (see our prompting tips guide) rather than on guessing which model should answer it.