Perplexity
Decider Flash V1.2
perplexity/pplx-decider-flash-v1.2
Decider Flash V1.2 is a 4B-parameter decision model from Perplexity, a smaller, lower-cost sibling of Decider V1.2 with the same API contract. Instead of generating text, it reads content passed...
RoleReturns probabilities for your questions, not a paragraph.
$0.010
In / M
Free
Out / M
262K
Context
- AA intelligence index
- Not on listing
- Inputs
- text, image
- Outputs
- decisions
- Tools
- Not listed
- Structured output
- Not listed
- Reasoning
- Not listed
- Weights
- Open · perplexity-ai/pplx-decider-flash-v1.2
- Listed
- Oct 16, 2026
01 · What it's good at
Use it for this.
Three jobs this model is built for, and the ones to give to something else.
01
Routing, moderation, grading, and scoring inside software
Skip it for
- Writing, chat, code, or open-ended explanations
02 · Example tasks
What you'd actually send it.
Real inputs, not a benchmark. Paste one into the snippet below to try it.
You send
State: ticket text. Questions: noul 'needs a human', choice of team.
Why this model
Decision models return probabilities.
03 · Your stack
Where it sits: Gate & route.
You send content and typed questions (yes/no, pick one, score a rubric). You get probabilities. Your code applies a threshold.
Layer 01
Retrieve
Layer 02
Plan & reason
Layer 03
Do the volume
Layer 04 · this model
Gate & route
Layer 05
Draw, speak, listen
04 · Wire it up
One request. Paste and run.
Uses OpenRouter, so one key works for every model on this site. Set OPENROUTER_API_KEY and swap the prompt for one of the tasks above.
const res = await fetch("https://openrouter.ai/api/alpha/decisions", {
method: "POST",
headers: {
Authorization: `Bearer ${process.env.OPENROUTER_API_KEY}`,
"Content-Type": "application/json",
},
body: JSON.stringify({
"model": "perplexity/pplx-decider-flash-v1.2",
"state": {
"ticket": "I was charged twice this month."
},
"questions": {
"needs_human": {
"type": "noul",
"instructions": "Does this ticket need a human?"
}
}
}),
});
if (!res.ok) throw new Error(`Decisions ${res.status}: ${await res.text()}`);
const { answers } = await res.json();
// A noul answer is the probability of "yes". Your code picks the thresholds.
const p = answers.needs_human.noul;
if (p >= 0.9) console.log("yes");
else if (p <= 0.1) console.log("no");
else console.log("send to a human", p);
Decisions is an alpha endpoint on OpenRouter. The TypeScript SDK exposes it as openRouter.alpha.decisions.create().
05 · Watch out for
Gotchas.
This lesson is generated from the model type and the live listing. It has not been hand-reviewed yet.
Check the listing for supported parameters before you rely on tools or structured output.
From our review · 2026-10-18
Flash uses the Decisions API. You send state plus noul, choice, or score questions and read probabilities, then threshold them in your code. The listing describes a 4B model with the same question types as Decider v1.2. Perplexity's docs say to test Flash on your labeled data and keep it only if it meets your accuracy bar.
Sources
- OpenRouter model listing ↗
- OpenRouter quickstart ↗
- Perplexity Decisions quickstart ↗
- OpenRouter model listing ↗
Lessons are written from each model's listing and the linked docs. We don't publish our own benchmark numbers; the index above is Artificial Analysis, as shown on OpenRouter.