Æ

AAIOS-TCM Console

multi-model cost & time router · live ledger
Workspace skills, history & tools are sandboxed to the active domain
Requests
—
Spent
—
Saved
—
Saved %
—
Cache hits
—
What do you want to know?
Ask in plain words. It picks the cheapest model good enough, and the receipt is in the ledger.

Sign in to ask, and to keep what you ask.

Use an API key instead

Create one on the server with bash ~/aaios-tcm/deploy/native/ctl.sh key console

Chat
Ask anything.
One chat, three efforts — Standard smart-routes one model, Council debates it, Super Heavy runs the full swarm. Reply to steer; it auto-compacts like Claude.
What the switches do
  • Clarify first — a cheap model reads your question first, sets its difficulty & tier, and asks you to clarify if it's vague. Better routing, tiny overhead.
  • Compound — finds your best prior answer to a similar question and builds on it, so your answers compound and improve every time you ask. Free & local. (On by default.)
  • Web search — grounds the answer in live web results: a cheap reader searches (news index too, when the question is about this week) and distills, and whatever model answers reads that digest. Works with every model, pinned or not; never cached. (Auto by default.)
Delete this conversation?Delete it?
IdleSay something, or type below.
Folders

Categories

—
History

History

Load a key, then ask something to start your history.
Registry

Models

ModelTierPrice in / out, per M tokensHealth
—

Websites

Sign in to see the sites, pages and logins this workspace holds.

How savings work

Economics

How the saving is computed

Each request is priced at the model that answered and again at the baseline model. The gap is your saving, and four levers create it.

1
Routing. A cheap classifier scores how hard each prompt is and sends it to the cheapest model that is good enough. Most of the saving comes from here.
2
Exact cache. The identical prompt, asked again, comes back instantly for $0.
3
Semantic cache. A reworded past question, in any language, also comes back for $0.
4
Batch lane. Work that can wait an hour or more goes through the providers' batch APIs at half price.
Hands-on

See each lever fire

A
Ask a question in Chat and note the saved line under the answer. That is routing.
B
Send the same prompt again. It comes back as exact_hit and costs $0.
C
Reword it and send it. It comes back as semantic_hit and costs $0.
D
Turn on Clarify first. A cheap model sizes the question, escalates only when it truly needs a bigger model, and asks you to clarify when it is vague.