Editorial illustration for Perplexity's 19‑model AI ‘Computer’ debut; Anthropic lets fan‑fav Opus 3 blog
Perplexity Computer: AI Super Agent with 19 Top Models
Perplexity's 19‑model AI ‘Computer’ debut; Anthropic lets fan‑fav Opus 3 blog
Perplexity just rolled out its so‑called “Computer,” a suite that stitches together 19 distinct language models under one interface. The launch, announced in a terse blog post, promises users the ability to tap different specialized engines without juggling separate accounts or APIs. At the same time, Anthropic has taken an unexpected step: it’s kept one of its retired models, Opus 3, online long enough to author its own blog entry.
The move is more than a nostalgic nod; it surfaces a debate that’s been simmering among AI developers about how to treat models that are no longer in production. By giving Opus 3 a voice after retirement, Anthropic signals a willingness to explore the boundaries of machine agency while also positioning itself against a competitor that’s currently navigating a “hornet’s nest” of public scrutiny. The juxtaposition of Perplexity’s multi‑model rollout and Anthropic’s unconventional preservation of a fan‑favorite sets the stage for a deeper look at why this matters.
Why it matters: Opus 3 was a fan favorite of a model, and Anthropic preserving it (and even letting it write a blog) in retirement is both in line with how it has prioritized exploring AI consciousness and welfare, and also an easy PR win over OpenAI -- which is still dealing with a hornet's nest of users angry about the removal of its 4o model. AI TRAINING The Rundown: In this guide, you'll turn all those bookmarked articles you saved for "later" into something useful. You'll set up Perplexity's Comet browser to read through the articles, score each one by usefulness, and log the best finds into a Google Sheet.
Is the 19‑model “Computer” the next step for background AI agents? Perplexity says its new stack stitches together work from frontier labs and internal talent, promising a single workflow that can juggle many tasks. The claim rests on the premise that users, after OpenClaw’s reception, desire agents that operate quietly yet effectively. Yet the article offers no performance data, leaving it unclear whether the added model variety translates into measurable gains.
Meanwhile, Anthropic’s decision to keep Opus 3 alive—letting the retired model author a blog post—signals a continued focus on AI consciousness and welfare. That move also doubles as a public‑relations counterpoint to OpenAI, which the piece hints is wrestling with its own challenges. Whether preserving a fan‑favorite model will influence broader research agendas remains uncertain.
Both companies appear to be testing the limits of agentic ambition, but the real impact on everyday users is still ambiguous. As the AI community watches these experiments unfold, the practical benefits of such multi‑model orchestration and nostalgic model releases have yet to be demonstrated.
Further Reading
- Perplexity Computer: Unified Multi-Agent Autonomous AI Platform - Solid Timing
- Perplexity Computer turns AI into a full-time worker: How it works - Firstpost
- Perplexity debuts 'Computer' as AI search challenger pivots to ... - Hindustan Times
- Perplexity Computer Unveiled as Unified, Multi-Model AI Workflow ... - Gadgets 360
Related Reading
Browse all Anthropic news →
Business & Startups OpenAI, a Series F San Francisco startup founded in 2015 by eight pioneers
Business & Startups Terminal-Bench 2.0 launches with Harbor, testing any container-installable agent
Business & Startups Zuckerberg Unveils Meta Compute to Build Global AI Infrastructure
Related Topic Anthropic's Claude also citing Elon Musk's Grokipedia, reports say
LLMs & Generative AI
Related Topic Claude Code Update Boosts Context Window for Complex Developer Workflows
LLMs & Generative AI
From Archives Gong unveils Mission Andromeda AI coaching, chatbot and open MCP links
![Anthropic logo next to DeepSeek logo, symbolizing AI intellectual property dispute and copying [futurism.com].](https://cms.aidailypost.com/uploads/anthropic_calls_out_ai_copycats_tely_ai_offers_1week_content_lower_7df572975c.webp)
From Archives Anthropic calls out AI copycats; Tely AI offers 1‑week content at lower cost
From Archives Anthropic links Claude's psychological security and sense of self to its safety
LLMs & Generative AI
From Archives Pentagon’s Claude platform cuts costs 70% with secure single‑tenant AI
LLMs & Generative AI
Common Questions Answered
How does Perplexity's new 'Computer' differ from other AI systems?
Perplexity's Computer is a multiagent orchestration system that harnesses capabilities from over a dozen frontier AI models. It works like a CEO, delegating tasks across different specialized AI models and breaking down complex tasks into subtasks automatically.
What makes Perplexity's Computer unique in the AI agent landscape?
Computer is positioned as a safer alternative to OpenClaw, with the ability to reason, delegate, search, build, remember, code, and deliver tasks. It currently is available only to Perplexity Max users, with plans to expand to Enterprise and Pro subscribers in the coming weeks.
Why did Perplexity develop a multimodel AI system?
Perplexity believes that AI models have specialized capabilities rather than being truly general-purpose tools. By combining multiple models, Computer can approach complex tasks more effectively, similar to how a company uses different teams with specialized skills to complete projects.