Not every request needs the most powerful model. Here's how our intelligent routing system picks the right Claude model for every task.
Related: How the We.Inc AI website builder works, with real sites it built
Every AI request is different. A one-word color tweak and a full multi-page app do not need the same firepower. So instead of sending every message to the same model, We.Inc routes each task to the right AI model for the job.
Here's why, and how it works.
If you send every request to the single most powerful model, two things go wrong:
Match the model to the task, and you get fast responses on the easy stuff and deep reasoning on the hard stuff.
Our system has three layers:
When you send a message, a lightweight classifier categorizes it in under 200ms:
For unambiguous prompts, we skip the classifier entirely and use heuristics: pattern matching on keywords and message length. This saves a round-trip for ~60% of messages.
Based on the task complexity, we route to the right Claude tier:
| Task Type | Model tier | Why |
|---|---|---|
| Questions, simple edits | Fast | Sub-second, low cost, good enough |
| Standard pages, components | Balanced | Best balance of speed and quality |
| Complex architecture, multi-file | Most capable | Deepest reasoning, worth the cost |
Pro plan users get access to all three tiers. Starter users get the fast and balanced tiers.
If a generation fails (timeout, rate limit, transient error), We.Inc automatically retries so the request still completes.
The user never sees an error. A retry adds 1-2 seconds of latency, but the generation finishes successfully. In practice, retries trigger only a small fraction of the time: rare enough not to affect quality, frequent enough to matter for reliability.
Different model tiers cost different amounts. Instead of per-message pricing, we use credits:
The free plan has no AI credits: it lets you build up to 3 sites from templates and publish them free on a we.inc address. AI building and editing start on Starter at $20/mo (250 credits).
After every generation, we run a verification loop:
If verification fails, the system automatically attempts a fix before showing you the result. You only see the final, working output.
Most users don't care which AI model generates their website. But they care about:
The routing system makes all of this invisible. You just type what you want and get a great result. That's the point.
We.Inc is powered by advanced AI models, with fast, balanced, and most-capable tiers for different kinds of work.
Pro plan users can select model tiers (Core, Pro, Apex). The system also auto-routes based on task complexity.
We.Inc is an AI-powered website builder you can resell under your own brand. Launch a branded client dashboard, bill on Stripe Connect, and deliver AI-generated websites in minutes. White-label plans start at $99 a month for 25 client sites, with a 7-day free trial and no per-site fees.