Perplexity subscribers are set to gain access to a powerful new agentic tool this week, signaling a significant evolution in the company’s offerings.
Dubbed Perplexity Computer, the company describes it as a system that “unifies every current AI capability into a single system.” More specifically, this computer user agent is designed to autonomously execute intricate workflows, leveraging 19 distinct AI models and even generating subagents to tackle specific problems.
This advanced tool is exclusively available on Perplexity’s premium subscription tier, Perplexity Max, priced at $200 per month. Operating entirely in the cloud, it may circumvent some of the security concerns associated with other agentic tools like OpenClaw.
While a hands-on demonstration has not yet been conducted by TechCrunch, example workflows displayed on Perplexity’s website showcase its ability to manage tasks such as gathering statistical, financial, or legal data, performing in-depth analysis, and presenting findings as polished websites or visualizations.
Perplexity executives recently held a background briefing with the press to discuss the product and outline their annual strategy. A planned demonstration of the tool was unfortunately canceled just hours before the event, attributed to the discovery of product flaws.
The introduction of Perplexity Computer marks a new chapter for Perplexity, a company that initially garnered attention during the early AI boom by integrating cutting-edge models into user-friendly interfaces, particularly its search-engine-like answer service. This was followed by the launch of its Comet web browser last summer. An executive noted that competitors, including Google, have since adapted their products to resemble Perplexity’s offerings, a development seen as both a compliment and a competitive challenge.
The company is strategically adapting to a changing ecosystem. After being among the first AI firms to offer advertising, Perplexity discontinued this business late last year, explaining last week that it compromised users' trust in the accuracy of their answers. However, Perplexity’s user base, numbering in the tens of millions, remains significantly smaller than OpenAI’s, which boasts 800 million weekly users and began testing ads in ChatGPT this year.
Perplexity executives now articulate a focus on a more specialized user base, targeting individuals involved in “GDP-moving decisions.” During the briefing, executives, who requested anonymity, indicated a prioritization of enterprise subscriptions, especially for deep research applications.
“You don’t hear us talk about MAUs ever, because we’re not actually on a mission to get as many users as possible,” one executive stated, reinforcing this strategic pivot.
In a related development, Perplexity recently unveiled a new benchmark for complex research tasks, named Draco, where its own deep research offering predictably outperformed rivals like Gemini.
Perplexity asserts its independence from external APIs for its web index, now utilizing its own AI-optimized search API. The company is intensifying its efforts to package frontier models within a consumer-friendly user experience, emphasizing the value of orchestrating multiple third-party large language models (LLMs) to deliver the most cost-effective and precise answers to queries.
“Multi-model is the future,” argued a Perplexity executive, positing that models are specializing rather than commoditizing. The company has observed that its users frequently switch between models to achieve desired results; for instance, December 2025 queries for visual outputs were predominantly sent to Gemini Flash, software engineering tasks to Claude Sonnet 4.5, and medical research to GPT-5.1.
Perplexity’s software intelligently selects the optimal LLM for a given task, for example, choosing one adept at coding over another more skilled at drafting marketing copy. Executives also cited the use of Perplexity’s modified open-source Chinese-built LLMs for more economical query responses—a practice that drew criticism last year for lack of transparency, but one that, if conducted openly, could efficiently optimize LLM queries.
Furthermore, the company introduces "Model Council," a feature allowing users to query multiple models simultaneously. Nevertheless, the unit economics of providing multiple queries at flat subscription rates are not yet fully clear.
Despite these considerations, Perplexity maintains that its strategy of allocating tokens to the most suitable model for each purpose, coupled with high margins on user fees and the absence of costly infrastructure projects, will sustain its competitive edge, according to the executives.
Looking ahead, Perplexity has several initiatives planned: the Comet browser is slated for an iOS launch next month, and the company will host a developers’ conference, "Ask," on March 11 in San Francisco, aimed at promoting third-party utilization of its API.
One executive remarked on a shift in internal metrics, noting he now prioritizes recent revenue figures over daily query counts. This apparent focus on the bottom line has been observed by some customers, with the Perplexity subreddit frequently featuring complaints about new rate limits on both free and subscription tiers.
However, executives at the briefing unequivocally dismissed these concerns. “Any discussions on the free tier being made worse or rate-limited is completely false,” an executive affirmed.
The Editorial Staff at AIChief is a team of professional content writers with extensive experience in AI and marketing. Founded in 2025, AIChief has quickly grown into the largest free AI resource hub in the industry.