I'll admit I didn't think much about this for a long time. The "smart" cloud models answered my questions, drafted my emails, and helped me think things through. My week kept moving. Nothing on the surface screamed that I was wasting money. But under the hood, I was sending everything — including "thanks" and "what's for dinner?" — to a brain priced for hard work.

If you're just getting into AI, or you're weighing whether any of this is worth a quiet evening after the kids are down, this change matters more than the product names suggest. The jump isn't about becoming an engineer overnight. It's about putting a traffic director between you and a pile of models so easy questions stay cheap and hard ones can cost more — on purpose.

A model router is a helpful friend, not a new religion

You keep chatting. It picks the sensible brain behind the scenes.

Skip the buzzwords if you hate them. Day to day, a model router sits between you and a bunch of AIs.

You keep talking in your usual app. Behind the scenes, that friend peeks at what you asked and picks a sensible option:

  • Easy stuff → a model on your computer (often free after setup)
  • Hard stuff → a stronger cloud model (the one you pay for by how much text you send)

You don't have to keep opening settings and switching models. You just talk.

Trash Panda's version of that friend is a small free tool that runs on your PC. Open source lives here: github.com/jkhall316/trashpanda-router

Why bother when the default already "works"

Sports-car rates for short trips to the store add up

Cloud AI usually charges by tokens — little bits of text going in and out. Big models cost more per bit.

If every tiny chat hits a premium cloud model, you're paying sports-car prices for grocery runs. Saying thanks doesn't need the same brain as "help me rebuild this whole app."

A router helps you:

  • Save money on casual chat
  • Still get power when the job is real work
  • Stay in one place — same chat window, less fiddling

It's less "new AI product" and more "smart traffic director." Same habit as finding the model sweet spot: clear the bar for the job, then stop overbuying.

How you actually use it (high level)

Docker once, point your app once, then chat normally

You're not watching a sci-fi robot take over the laptop. You're doing something closer to this:

  1. Install Docker Desktop (it runs little apps in boxes — don't overthink it).
  2. Get the Trash Panda router from GitHub and start it (the README walks you through it).
  3. Open the simple admin page in your browser and set a couple basics (how much graphics memory you have, and cloud keys if you want cloud).
  4. In your chat app, point it at the router once and pick the model called trashpanda/auto.
  5. Chat normally.

That's it. Your keys stay on your machine. This website only helps with which model looks like a good fit — it doesn't take over your accounts.

Works with common chat apps and coding helpers that can use a custom "OpenAI-style" connection (Open WebUI, OpenCode, and friends). Details live in the GitHub README — this page is just the why and the vibe.

Don't want to install anything yet?

Totally fine. Poke the clicky tools first.

Stay in ChatGPT or Cursor if that's all you have energy for. Seriously. This idea can wait on the shelf until one quiet evening.

When you're curious without installing, use the tools on this site first:

When you're ready for "just chat and let it choose," come back to the router.

What we're not doing tonight

No shopping list. No "best AI of 2026." No guilt for staying on defaults.

You don't need to become an engineer to use AI well. You need habits that don't waste money or attention.

A model router is one of those habits: easy questions stay cheap, hard questions can cost more — on purpose.

What's next

Skim the GitHub README, or just pick a model first

When you're ready, skim the router on GitHub. If you're still exploring, poke around Pick a model instead.

If you've been holding off because the AI internet sounds like a full-time job, start here instead. The win isn't a new identity. It's that your week still works — and when you do pay for a bigger brain, you meant to.