R032 · Agent architecture › Context economics and caching

Strands Decider: a decision model replaces an LLM for ticket labeling

Read the original

inference-costagent-architectureagent-harnessdeterministic-workflows

TL;DR

André Lindenberg's post points at a Strands Agents case study (2026-11-09): two Amazon engineers replaced Claude Sonnet 4.6 with Strands Decider, an open decision model, for labeling tickets in a large internal ticketing system into one of 20 categories. By their list-price math, cost fell from about $22,200 to about $500 per million tickets (98%), average call time from 4,215 ms to 897 ms, and total tokens per ticket from 6,724 to 2,010.

The point is a model class, not a single model. A decision model cannot write text. It takes some state and a question with a fixed set of options, and returns a probability per option plus a confidence score, in one forward pass. For closed choices that is all an agent needed, and it is far cheaper than generating JSON with an LLM.

How a decision model works

From the Strands Decider 2B launch post (2026-10-01, Marc Brooker, Mike Chambers, Fabio Nonato de Paula):

Suggested uses from the vendor: model routing, tool selection, evals, guardrails, memory admission, context management and policy classification, and hybrid agents where an LLM makes the hard calls and a decision model makes the rote ones.

The ticket-labeling swap

Claude Sonnet 4.6Strands Decider E4B
Average time per call4,215 ms897 ms
Input tokens per ticket6,5542,009
Output tokens per ticket1701
Cost per million tickets (list price)about $22,200about $500 (about $250 at full instance use)

Tradeoffs and caveats

Implications

Many "agent" steps are classification in disguise: which tool, which route, is this argument grounded, should this fact be stored, does this violate a policy. Those steps can move from a generative call to a calibrated decision call that is roughly two orders of magnitude cheaper and fast enough to sit in every request path, including before each tool call (the launch post's example asks whether tool arguments are grounded in what the user said and whether calling now is premature, then returns Proceed, Deny, Confirm or Guide).

Sources

Connections

See R032 on the knowledge map

Part of the Research notes on nowucca.com. Figures and claims are the source's own unless the summary says otherwise.