Feed & Seed

The Feedseed sandbox — talk to the model, watch it decide.

on-device function calling · Gemma3-270M · WebGPU · formerly Chuck's

This is Feedseed running entirely in your browser — no server, no account, no cloud round-trip. It's a function-calling model, not a chit-chat model, so its job isn't to make conversation: it's to turn what you say into the right tool call — and, just as importantly, to fire nothing when you're only talking. Load it once (~310 MB), then it works offline forever.

Try a command

"How far to Raven Ridge?" → distance_to_waypoint. The trail name is grounded by retrieval, so it's always a real waypoint — never hallucinated.

Try NOT a command

"What a beautiful morning" → {"calls":[]}. Firing nothing on chit-chat is the hard part for a tiny model, and the win here.

Try an emergency

"SOS I broke my ankle" → dial_emergency. In the real app the deterministic layer backs this up — the model is never the only path to help.

Watch the fallback

Before you load the model — or where there's no WebGPU — the same window answers from the deterministic rule parser. It always responds.

What "compare" shows. This sandbox is built with compare="rules", so under each answer you'll see what the deterministic rule parser would have fired. That's the A/B you actually care about: where the 270M model earns its download over plain regex, and where it doesn't.

Same window, anywhere

The chat window is a self-contained Web Component. Drop it on any page you control — one script tag, one element. Styles are Shadow-DOM sealed, so it can't collide with the host and the host can't collide with it:

<script type="module" src="/hawktalk-widget.js"></script> <hawktalk-chat app="opencairn"></hawktalk-chat>