I Route Every AI Agent to the Cheapest Model with Quota. Here's What Broke A developer built a small Python router that assigns each AI agent task to the cheapest model with remaining quota, using a tier list (small/medium/large) that keeps model names out of role prompts, a quota_status() check that drops any model above 90% usage, and a startup-failure detector that rolls to the next candidate within 60 seconds. The router deliberately never auto-escalates a task to a larger model, exiting loudly with skip reasons instead, after cheap models produced an infinite test loop that consumed 9 GB of RAM and 419 lines of duplicated helpers. The developer reports no dollar figure for the savings. I wanted coding with AI agents to be genuinely cheap, so I built a tiny Python router that sends every task to the cheapest model that still has quota left. Within three days, a cheap model wrote a test that looped forever and ate 9 GB of RAM, another stopped halfway with "Free usage limit reached", and a third dumped 419 lines of duplicated helper functions into a script because it assumed "the grader supplies tests." 😅 Everyone is asking whether coding with AI is actually faster and cheaper. My honest answer? It's fast and nearly free, right up until a cheap model writes something that takes two hours to unwind. So here's the router, the rules that came out of those crashes, and why it's forbidden from upgrading itself. TL;DR: The first mistake I made was tying each agent prompt to a model. Now the role "writer", "designer", "coder" is just a prompt with rules and skills, and a role never names a model . It only says what size of work it usually does. Why does that matter? Because the moment a model is hardcoded, you can't route around outages, renamed models or an exhausted rate limit. The tier list does that for you: { "small": "free-model-a", "free-model-b", "cheap-model-c" , "medium": "mid-model-a", "mid-model-b" , "large": "big-model-a", "big-model-b" } Cheapest first, more capable after. Swapping a model is a one-line change, and no prompt has to know. When a task comes in, the router walks the requested tier and filters it: python def select candidates tier, tiers, allow paid=False, forced=None : if forced: return forced , to try, skipped = , for cand in tiers tier : if cand.get "paid" and not allow paid: skipped.append cand, "paid model not allowed" continue if q := quota status cand : skipped.append cand, f"quota: {q}" continue to try.append cand return to try, skipped quota status reads the CLI's local usage numbers instead of hitting a rate-limit API. If any window rolling hours or weekly is above 90% , that model is out. Why 90 and not 100? Start a big task at 96% and it tends to run dry mid-stream, leaving half-written files behind. The buffer is cheaper than the cleanup. And yes, you can force a model by hand. The router just warns you and gets out of the way. What about a model that gets picked and then falls over before doing anything? AI CLIs fail at startup in creative ways: renamed models, instant 429s, free tiers that cut you off at second zero. So the router watches the first minute: STARTUP ERRORS = "429", "model not found", "Free usage limit", "quota" def is startup failure code, seconds, log : if seconds 60: return False if any e.lower in log.lower for e in STARTUP ERRORS : return True return code = 0 Fail fast, and it rolls to the next candidate in the same tier. Every run also leaves its prompt, raw log, final answer and the choice it made, so I can see why a model was skipped without guessing. Did the router make everything smooth? Not even close. A few of my favourite disasters: model not found . Lesson: only route to models you've actually run. If you like this kind of thing, my previous post on running a side project with AI agents https://dev.to/galadev/i-run-a-side-business-with-ai-agents-heres-what-broke-in-14-days-4ch has more. When every small model fails, why not bump the task to a bigger one automatically? Because the router never decides that a task is bigger. Only I do. When a small task fails across the board, it's rarely the model being dumb. It's usually: Auto-escalating just pays premium prices to retry a broken prompt. So when a tier runs out, the router exits loudly and lists why it skipped each model. Then I either fix the prompt or bump the size by hand. The machine executes; the human owns the scope. I don't have a dollar figure, since most of this runs on subscriptions and free tiers rather than metered APIs. What it saved was headroom and attention : The one rule I'd keep if I had to delete everything else: Cheapest-with-quota is the right default for writing and small scoped edits. For anything with tests, set a higher floor. Saving ten minutes of quota is worthless if you spend an hour reverting a cheap model's "tests". I use this router to keep a few small data sites running on my own, like a cheapest fuel price map for Spain https://fuelspot.devetty.es/?utm source=devto&utm medium=article&utm campaign=cheapest-model-router&utm content=intro and a public tenders finder https://licitacionesalerta.es/?utm source=devto&utm medium=article&utm campaign=cheapest-model-router&utm content=outro . The agents are the easy part. Getting people to find and use the sites is the hard one. So, for anyone who's grown a side project: I read and reply to every comment 💛 I write these with help from AI agents.