# Think when unsure

> Have the server read an unsure question again after the model writes its reasoning, and spend the extra time only on those questions.

Source: https://opendecisio.com/docs/think/

The [questions](https://opendecisio.com/docs/glossary/#question) decisio is unsure of are the ones a single read most often gets wrong. With thinking on, each of them is read again after the served model writes its reasoning, to get more of them right. The extra time goes to those questions only: a sure question is still answered from one read, in milliseconds.

 

[Figure: Only an unsure question is read again, after the model writes its reasoning. Four steps. 1: every question is read once, as always. 2: it is unsure when its top probability is under its type's threshold, 0.75 for a choice, 0.92 for yes/no and 0.7 for a score, times the factor (1 by default), and it has at most 16 options; a sure question is answered from the one read. 3: an unsure question is read again after the served model writes up to 512 tokens of reasoning. 4: the two reads are mixed in log space and go through the same temperature.]

 

It follows the procedure of [Chon](https://github.com/xoai/chon) (Apache-2.0). No measurement of its accuracy or its time is published yet.

 

## 1 Turn it on

 

Start the server with `--think`:

 

```
pip install "decisio[serve]==0.22.0"
python -m decisio.serve.vllm_engine --think
```

 

Shown, not run: it needs Linux with one NVIDIA card, which this check does not have.

 

- Thinking is off by default, on every base: a request is one read, as fast and as repeatable as before.
- `--no-think` turns it off, including when a tuned model’s repository turns it on in its `defaults.think`.
- A request decides for itself with `"think": true` or `"think": false`; without the field, the server’s setting applies.

 

## 2 Ask with it

 

```
curl -s http://127.0.0.1:8000/v1/systemone -H 'Content-Type: application/json' -d '{
  "state": "Our invoice shows two seats, but we removed the second user last month.",
  "think": true,
  "questions": {"queue": {"type": "choice", "instructions": "Which team should handle this ticket?",
    "criteria": {"billing": "Invoices and charges", "access": "Logins and seats", "bug": "The product misbehaves"}}}
}' | python3 -c 'import json, sys; r = json.load(sys.stdin); print(json.dumps({"thinking": r.get("thinking"), "thinking_tokens": r["usage"].get("thinking_tokens")}, indent=1))'
```

 

Shown, not run: it needs a server that thinks; the CPU stand-in cannot write reasoning, only the vLLM engine on a card can.

 

- A question thinks when its [top probability](https://opendecisio.com/docs/glossary/#top-probability) is under its type’s [threshold](https://opendecisio.com/docs/glossary/#threshold): 0.75 for a choice, 0.92 for yes/no, 0.70 for a score.
- `--think-factor` scales the three thresholds, 1 by default; a lower factor thinks on fewer questions.
- `--think-budget` caps the reasoning at 512 tokens per question by default, from 16 to 4,096. It is set on the server only.
- All the unsure questions of a request go to the model in one call.

 

## 3 Read what it did

 

- `thinking` in the response says how many questions thought, the reasoning tokens and the time, and for each question whether it thought, or why not.
- `usage.thinking_tokens` counts the reasoning tokens, and `usage.output_tokens` includes them.
- The headers `x-decisio-think-ms` and `x-decisio-think-tokens` carry the time and the tokens.

 

The server’s setting, and how many questions have thought, are in `/health`:

 

```
curl -s http://127.0.0.1:8000/health \
  | python3 -c 'import json, sys; print(json.dumps(json.load(sys.stdin)["systemone"]["think"], indent=1))'
```

 

Run against the CPU stand-in at v0.22.0 on 2026-10-21. It printed:

```
{
 "default": false,
 "factor": 1.0,
 "budget": 512,
 "requests": 0,
 "questions": 0,
 "tokens": 0,
 "ms": 0.0
}
```

 

## 4 Know what stays at one read

 

- A question with more than 16 [options](https://opendecisio.com/docs/glossary/#option).
- A question a registered [task](https://opendecisio.com/docs/glossary/#task) answers: its correction and certificate were fitted on one read.
- Every question of a request with images, an [abstention](https://opendecisio.com/docs/glossary/#abstain) option or two orders.

 

The response says why a question did not think. Where it runs, how it shares the server and what it does to certificates are in [Requirements and limits](https://opendecisio.com/docs/limits/#3-set-up-for-production).

 

## What next

 

- [Requirements and limits](https://opendecisio.com/docs/limits/#3-set-up-for-production): thinking in production.
- [Server flags](https://opendecisio.com/docs/flags/#every-flag): `--think`, `--think-factor` and `--think-budget` in full.
- [The HTTP API](https://opendecisio.com/docs/api/): the `think` field and the `thinking` answer.