product updatePerplexity AI

Perplexity launches Monitor API: scheduled web search that pushes changes to agents via webhook, $1 per 1,000 runs

TL;DR

Perplexity has launched the Monitor API, which searches the web on a schedule and sends only relevant changes to apps and agents via webhook. Web monitors cost $1 per 1,000 successful runs, and page monitors cost $1 per 1,000 fetched URLs. In an early internal test, Perplexity claims it delivered 63 fresh, relevant pages in eight hours against 20 for Firecrawl.

3 min read
0

Perplexity has released the Monitor API, a service that runs web searches on a schedule and sends relevant changes to apps and AI agents via webhook. It is available now through the API and the API console, priced at $1 per 1,000 requests, according to The Decoder.

How it works

A single API request creates a monitor by setting the search query, goal, schedule, and webhook URL. Perplexity says that until now, engineering teams had to trigger recurring searches themselves and maintain custom code to remove duplicates and detect changes. The company says this becomes nearly impossible to maintain across hundreds of topics and pages.

The API has two modes:

  • Web monitors run a search query through Perplexity's web and news search on a regular schedule, for example to catch new funding rounds at AI infrastructure companies.
  • Page monitors track up to 10 pages, such as pricing pages, changelogs, or policies. The first run stores a baseline version, and later runs return a text diff of any changes.

Schedules range from every 15 minutes to weekly, or a custom cron schedule. Each run picks up where the last one left off, and the service filters out duplicate results. Webhooks are signed, and Perplexity automatically retries delivery after temporary errors. Developers can pause, resume, or manually trigger monitors through the API or the console.

Decision model filters results

Perplexity evaluates results with its own decision model against a goal that developers write in natural language. Web search results must clear checks for relevance, source credibility, novelty, and timeliness. Page changes are judged on whether they matter to the goal. An update with a confidence score is sent only when every threshold is met.

Developers can adjust the thresholds for each check individually and change the goal without recreating the monitor. In one example from Perplexity, the decision model filters out previews, rumors, and recaps and forwards only genuine Series A and Series B announcements.

Pricing and access

  • Web monitors: $1 per 1,000 successful runs
  • Page monitors: $1 per 1,000 fetched URLs
  • The Monitor API uses the same key as the Search, Context, and Agent APIs

Benchmark claims

Perplexity tested the Monitor API against monitoring services from Firecrawl, Exa, and Parallel across 60 topics drawn from real Search API queries, covering company news, tech releases, and regulation. Each provider ran with its own schedules and settings, and a separate model scored each result on relevance and recency.

In the first eight hours, Perplexity says it delivered 63 fresh, relevant pages. Firecrawl delivered 20, Exa 15, and Parallel Core and Lite 10 each. Perplexity describes these as early internal results. The same three competitors led an August Artificial Analysis benchmark of traditional search APIs for AI agents, but Perplexity was not included in that test.

Perplexity points to use cases including sales teams tracking funding rounds and staffing changes at target accounts, analysts watching quarterly earnings and acquisitions, agents that flag competitor price changes, and agents that summarize new regulatory rules for compliance teams.

What this means

The Monitor API moves recurring-search logic, including deduplication, change detection, and relevance filtering, from developer code into a hosted service. Agents receive a webhook only when every configured threshold is met, so they do not need to poll or process unfiltered results. Per-check thresholds and an editable goal let developers tune sensitivity without rebuilding monitors.

The benchmark figures come from Perplexity's own test, scored by a separate model, and the company labels them early. They are a vendor claim, not an independent comparison. Because Perplexity was absent from the August Artificial Analysis benchmark, there is no third-party result that includes it alongside these services.

The launch adds to Perplexity's developer lineup, which already includes the Search, Context, and Agent APIs under a shared key. The Decoder notes the company is also betting on agents such as Computer and its always-on Teammates, along with developer tools like Search as Code.

Related Articles

Comments

Loading...