You never have to pick a model.
New AI models launch every few weeks, each with a new name and a new claim. Here’s what that means for you as a Jinn customer: nothing. You’ll never compare them, configure them, or fall behind on them. Every job in the suite is routed to the class of model it needs — fast and small where volume matters, the most capable class only where the work truly demands it — and Jinn keeps that routing current as vendors ship new families.
You’ll notice this page names classes of model, never brand-name models. That’s deliberate: the rules are written to outlive every launch, so what you read here stays true when next quarter’s model arrives.
What it means for you
You never choose, tune, or keep up with an AI model. Every job gets the class of model it needs, automatically.
How it decides
One routing table maps each job to a model class by plan tier. When a better model ships, the table updates once and the whole suite moves at once.
What it protects
Your money. High-volume jobs are pinned to fast, small models by rule, so a premium plan buys you more coverage — more checks, more engines — never waste on an oversized model.
Which model does which job.
The same truth, framed by what each job does for you. These are the jobs you’ll meet most; the table holds more for work inside individual products.
- monitoring
- Watching how AI engines talk about your brand, around the clock. These checks run constantly, so they run on fast, small models — by rule, not by accident. A premium plan buys you more checks across more engines, never a bigger model doing the same check.
- fidelity
- Judging whether an AI engine described your brand correctly. That judgment runs on a mid-weight model, and premium plans step it up to the most capable class.
- pin copy
- Short, high-volume copy, like a batch of 75 pins with several AI calls each. Fast, small models on every tier: short copy in your brand voice doesn’t need an expensive model, and a batch that size couldn’t afford one.
- deliverables
- Long-form strategy documents. Routine sections run on the workhorse class; the parts that have to think hardest escalate to the most capable model available. If a model errors mid-run, the call steps down one class with a logged warning — never a silent swap you’d never hear about.
- briefs
- Strategy briefs always run on the most capable class. They’re rare and you asked for them, so the budget holds even at the top.
Overrides are allowed, but every one is logged with a reason — the default path is where the guardrails live.
That is the machinery that picks the model. The output is brand work you never have to configure. See it on your brand
Every call, accountable.
Every AI call in the suite passes through one gateway. Routing decides which model; the gateway makes sure every call is accountable, bounded, and honest about cost. Here’s what that buys you:
- metered
- Every call records its own cost the moment it finishes. Nothing is generated off the books.
- ceilinged
- Budget is set aside before a call runs, then settled to the real cost after. If the meter that tracks spending is ever unreachable, the call is refused — a broken meter stops spending, it never allows spending that isn’t tracked.
- bounded
- Every call has a deadline. A stuck engine is cut off and handled — it never leaves your work hanging.
- typed
- Failures come back with names — rate limit, refusal, timeout — so the product responds properly instead of guessing.
- priced
- Four engine families, one price list: Anthropic, OpenAI, Google, Perplexity. A model Jinn doesn’t recognize is priced deliberately high and flagged — costs are never quietly under-counted.
Every call is metered; a call is marked as an additional cost only when a product opts in, which is what keeps the totals honest and impossible to double-count.
Why you never keep up with model launches.
When a vendor ships a new model family, the change is one entry in the routing table — shipped to every product at once. There’s no per-product upgrade to chase and no setting for you to touch. Your whole suite moves to the better model on the same day, and you find out because the work got better.
That’s the point of routing by class instead of by name: the rule "high-volume work runs on fast, small models" doesn’t care what this quarter’s fast, small model is called. You get the right model for every job, and the right model keeps getting better underneath you.
Keep exploring how Jinn works.
This page is the model layer. The full tour shows how it connects to everything Jinn learns, wears, and works for your brand.
- The full tourThe mechanism end to end: how Jinn learns, wears, and works a brand.
- The recordThe 346-signal Brand DNA record, group by group.
- Guardrails & spendThe fixed rules that guard your money and your name.
- Measurement disciplineThe bar our own numbers must clear before we show them to you.
- Check it yourselfWhat “verified” actually means here: human approvals, recorded history, and facts that age on purpose.
Models, answered.
- Do I choose which AI model Jinn uses?
- No — and that’s the feature. Every job is routed to the class of model it needs. You never pick a model, tune one, or keep up with launches.
- Why not use the most powerful model for everything?
- Because you’d be paying for it. The most powerful model on a high-volume path breaks the budget immediately, so those paths are pinned to fast, small models by rule. A premium plan buys more checks across more engines — never waste.
- What happens when a new model comes out?
- The routing table is one file, so a new family is one entry — shipped to every product at once. You get the upgrade without changing anything.
- How do you keep model spend under control?
- Every call passes through one gateway that records its cost, sets budget aside before the call, and settles it after. If the meter that tracks spending is unreachable, the call is refused — spending stops rather than going untracked.
The right model for every job, chosen for you.
Run the free brand read and watch the suite route every job to the model class it needs — without a single setting from you.