---
title: "Perplexity API key"
description: "Add your Perplexity API key for web-grounded AI share-of-voice probes: where to get the key, model options and costs, and troubleshooting."
source: https://backtalk.sh/docs/perplexity
---
# Perplexity key

Bring your own [Perplexity](https://www.perplexity.ai/) key to probe **Perplexity's web-grounded answers** for AI share of voice. Perplexity searches the live web before it answers, so its recommendations track what the internet currently says about your category; that makes it the most listening-adjacent surface BackTalk probes. Like any saved AI key, it can also serve as your filtering default, though the always-on web grounding makes it a better probe than scorer.

## What it unlocks

- Perplexity joins your AI share-of-voice probes: your topics' prompts are asked on schedule and every answer is measured for who got recommended.
- Web-grounded answers with **citations**: the sources Perplexity used are captured with each sampled answer.
- Optionally, filtering (sentiment, categories, priority) if you assign it as the signal-filtering model under **What runs each task**, like any provider.

You can save **a key per provider**. Left untouched, your first key powers filtering (sentiment, relevance, priority) and every saved key's model joins the AI share-of-voice comparison. Point either job at any saved key and model, including several models from one provider, under **What runs each task** in Settings.

## Get your API key

1. Open [**Perplexity → Settings → API**](https://www.perplexity.ai/settings/api) and sign in.
2. Add a payment method if you haven't (the API is pay-as-you-go), then choose **Generate key**.
3. Copy the key. It starts with `pplx-…`.

## Add it in BackTalk

1. Go to [**Settings**](/app/settings/ai) and find the **Perplexity** card under AI providers.
2. Pick a model (see below) and paste your `pplx-…` key.
3. **Save.** BackTalk sends one tiny test request to confirm the key and model, then stores the key encrypted.

## Models & cost

Billed by Perplexity directly at their token rates, with no markup from BackTalk. List prices (USD per million input / output tokens):

- **Sonar**: ~$1 / $1. Fast web-grounded answers; the right default for share-of-voice probes.
- **Sonar Pro**: ~$3 / $15. Deeper grounding and richer citations for harder questions.

Note that Perplexity also meters web searches per request on top of tokens; the Usage & cost page tracks the token side, and Perplexity's console is authoritative for the total.

## Troubleshooting

- **"The provider rejected that API key."** The key is wrong or was revoked. Generate a fresh one from the API settings page.
- **Quota / rate-limit errors.** Pay-as-you-go accounts have per-minute request limits; spread probes with a slower cadence or raise your limit with Perplexity.

> **Your key stays private.** It's encrypted at rest and only decrypted server-side when a probe or scoring call runs. It is never returned to your browser and never logged. Settings shows only a masked hint.
>
> **Rotating your key:** generate a new key in Perplexity's settings, paste it into Settings to replace the old one, then delete the previous key.