langgraph-decision-models

1<overview>
2A **decision model** answers typed questions about state and returns probabilities instead of prose. It replaces the common pattern of prompting an LLM, parsing its text, and branching on the result.
3 
4- **`Noul(instructions=...)`** — binary question, returns a probability of yes
5- **`Choice(instructions=..., criteria={...})`** — picks one label, returns the full distribution plus confidence
6- **`Score(instructions=..., criteria=[...])`** — grades against an ordered rubric, returns an expected value plus confidence
7 
8`TypeSafeClassifier` is a LangChain `Runnable[ClassifierRequest, ClassifierResponse]`, so it drops into a node like any other runnable. Up to 32 questions share one request and are answered independently — your code combines them.
9 
10**Reach for one when** a node generates text only so you can parse a decision out of it: routing, triage, filtering, guardrails, or per-item classification over a batch.
11 
12**Do not** reach for one when the node's output is the product (summaries, drafts, code) or when the judgment needs multi-step reasoning. A decision model classifies; it does not think.
13</overview>
14 
15---
16 
17## Install and wire
18 
19`langchain-typesafe` is alpha (`0.0.1a3`) and `TypeSafeClassifier` is marked `@beta` — pin it and expect churn.
20 
21```bash
22uv add langchain-typesafe
23```
24 
25Three ways to reach a model. The classifier POSTs to `{base_url}/v1/systemone` with `Authorization: Bearer {api_key}`, so switching providers is constructor arguments only:
26 
27<ex-wiring>
28…