Perplexity Decisions API moves the latest aliases to pplx-decider-v1.3 and adds locate questions (November 2026)
pplx-decider-latest now selects v1.3, so tuned thresholds can shift. New locate question type returns matching text spans. Prices unchanged.
Your keys
0 of 1 ready
Create each key on its own page, paste it below, then copy the finished lines into your project's .env file. What you paste stays in this tab. It is never sent to Wirestack or saved.
- Perplexity API keyPERPLEXITY_API_KEYGet key
Open the API Keys page in the Perplexity API console and generate a key. Add credits in Billing first.
What changed
In November 2026 (the changelog gives only the month) Perplexity released two new decision models for the Decisions API: pplx-decider-v1.3 (27B) and pplx-decider-flash-v1.3 (4B). Prices match v1.2: $0.02 and $0.01 per million input tokens, and output tokens stay free. Both accept text and images and answer several questions about the same state in one pass.
The aliases moved. pplx-decider-latest now selects pplx-decider-v1.3, and pplx-decider-flash-latest selects pplx-decider-flash-v1.3.
There is a new question type, locate. It returns the spans of a string state that match a description, with character offsets and a score for each span. The quickstart shows the question:
{
"type": "locate",
"instructions": "Every product name mentioned in the review."
}Each span in the answer has text, start, end (exclusive) and score. When nothing matches, spans is [].
What to do
If your integration calls pplx-decider-latest or pplx-decider-flash-latest, it is already on v1.3. Perplexity says probabilities can change with the model, so revalidate any thresholds you tuned on v1.2. To stay on v1.2 while you do, name pplx-decider-v1.2 or pplx-decider-flash-v1.2; both remain available.
For locate, the docs set these limits:
- Only the v1.3 models answer it; another model returns .
