Perplexity Launches ‘Computer’: A 19-Model AI System That Researches, Codes, Deploys, and Never Clocks Out

Key Takeaways

1. Perplexity AI has launched Perplexity Computer, a cloud-based agentic platform that orchestrates 19 AI models — including Claude Opus 4.6, GPT-5.2, Gemini, and Grok — to execute end-to-end projects autonomously.
2. Available now for Perplexity Max subscribers at $200/month, with 10,000 monthly credits and a one-time 20,000 bonus credit grant at launch. Perplexity plans to offer Pro for $20 per month and an Enterprise tier.
3. Scale of Interest: The announcement garnered 12 million views on X within 20 hours, signaling massive market appetite for multi-model agentic AI.​
4. Why It Matters: Computer is the first major platform to wire multi-model orchestration into an autonomous agent that can run for hours, weeks, or months — a direct challenge to single-model competitors like OpenAI’s Operator and Anthropic’s Claude Cowork.

Quick Recap

Perplexity AI officially unveiled Perplexity Computer on February 25, 2026, via its X account (@perplexity_ai), billing it as a system that “unifies every current AI capability into one system” capable of researching, designing, coding, deploying, and managing projects end-to-end. CEO Aravind Srinivas revealed the team had been building the product quietly for two months, calling it Perplexity’s “next big thing”. The tool is currently available to Max-tier subscribers and runs entirely in the cloud.

Inside Perplexity Computer: 19 Models, One Orchestrator

Perplexity Computer is not a chatbot upgrade, it is a multi-model orchestration platform that breaks complex projects into subtasks and, routes each one to whichever model is best suited for the job. Of the 19 models available, Perplexity has named six publicly:

Users describe an end goal, and Computer decomposes it into a dependency graph of subtasks, each running in an isolated sandbox with access to a real file system, a real browser, CLI tools, and hundreds of third-party connectors including Gmail, Slack, Notion, GitHub, Jira, and FactSet. The system can run autonomously for hours, weeks, or even months, delivering reports, deploying apps, and managing recurring workflows without constant supervision.

Perplexity employees have been dogfooding Computer since January. One team reportedly built a 4,000-row spreadsheet overnight that would normally take a week. As one Perplexity executive put it: “Multi-model is the future. Models are specializing, not commoditizing”.

Why Multi-Model Matters Now?

The launch arrives at a pivotal moment in the agentic AI race. OpenAI’s Operator, Anthropic’s Claude Cowork, and open-source projects like OpenClaw are all vying to turn AI from a conversational tool into an autonomous worker. But each is largely tied to a single provider’s models.

CEO Srinivas took a direct shot at Anthropic, writing that “the biggest weakness of Claude is that it only coworks with Claude”. Perplexity’s bet is that no single model excels at everything: its own data shows December 2025 users favored Gemini Flash for visual outputs, Claude Sonnet 4.5 for software engineering, and GPT-5.1 for medical research. Computer automates that model-switching behavior.

The broader market context is equally significant. Perplexity, valued at roughly $20 billion after a $200 million funding round in September 2025, is pivoting from a search-engine challenger into what it calls a platform for “GDP-moving decisions”. The company has abandoned advertising entirely, saying it undermined trust, and is focusing on enterprise and high-value individual subscribers.

Competitive Landscape

While Perplexity Computer leads in model diversity and long-running autonomy, OpenAI Operator offers the most accessible entry at $20/month with 40 agent tasks, making it more cost-effective for lightweight browser automation. Claude Cowork’s strength lies in deep desktop integration and Anthropic’s industry-leading coding models, but its single-provider lock-in limits flexibility compared to Computer’s 19-model roster.

Sci-Tech Today’s Takeaway

I think Perplexity just made the most ambitious move in the agentic AI space this year — and possibly the riskiest. In my experience covering AI launches, the products that try to “do everything” usually end up doing nothing particularly well. Computer’s multi-model architecture makes a genuinely different bet. Instead of asking, “Which AI is best?”, Perplexity argues that the question is outdated—because the answer is all of them, intelligently routed.

I’m cautiously bullish. The $200/month price tag with credit-based consumption isn’t cheap, but for professionals and teams whose workflows currently bounce across ChatGPT, Claude, Gemini, and a dozen SaaS tools, the consolidation alone could be worth it. The 12-million-view reception on X tells me the market is hungry for this.

My concern? Moat. If Computer’s value is orchestrating other companies’ models, what happens when OpenAI or Anthropic decide to build their own multi-model routers? Perplexity is betting its routing intelligence and integration layer are the defensible parts. I generally prefer companies that control their core technology end-to-end, but I’ll admit: nobody else has shipped anything like this at this scale. For now, Perplexity Computer is the product to watch — and for early adopters willing to invest $200/month, it could genuinely change how work gets done.