The latest AI news and updates of Perplexity — an AI-powered answer engine delivering real-time, cited answers to any question. Covering Perplexity's latest product updates, launches, and research from the past 90 days.
Perplexity19h agoPerplexity integrated Model Council into its Computer platform, enabling multi-model analysis for complex tasks. The tool evaluates a single issue across two to eight models from various providers. It synthesizes consensus, identifies disagreements, and generates work-ready assets like reports or decks based on the findings.
Perplexity19h agoPerplexity released Personal Computer for Windows, an always-on agent harness for Pro, Max, and Enterprise subscribers. The platform orchestrates 15+ models to perform tasks across local files, Microsoft 365 apps, and 400+ connected services. It features voice mode, Comet browser automation, and local folder access, allowing agents to read, edit, and manage work directly on the desktop.
PerplexityJul 25Perplexity added Anthropic’s Claude Opus 5 to its answer engine and Perplexity Computer platform. In internal WANDR benchmark evaluations, the model outperformed six other leading models, ranking second only to Fable 5. It delivers this performance at a 57% lower cost than Fable 5.
PerplexityJul 15Perplexity launched SPACE, a sandbox platform for long-running agentic workflows. SPACE provides secure, isolated environments for code execution and file management. It now powers 100% of Perplexity Computer traffic, reducing median sandbox creation latency from 185 milliseconds to 60 milliseconds and 90th-percentile latency from 447 milliseconds to 89 milliseconds.
PerplexityJul 14Perplexity open-sourced WANDR, an evaluation benchmark for research agents that must perform wide-and-deep knowledge work. The benchmark includes 500 realistic tasks requiring 170,495 source-backed records. WANDR grades agents by re-fetching cited pages and verifying claims against evidence, providing a dense signal to identify failures in discovery, enrichment, and factual accuracy across complex research hierarchies.
PerplexityJul 10Perplexity added xAI's Grok 4.5 as an orchestrator model for its Computer platform, available to Pro and Max subscribers. In WANDR benchmark evaluations, the model outperformed five other orchestrator configurations while costing roughly half as much as Opus 4.8.
PerplexityJul 10Perplexity launched Computer Analytics, a dashboard for tracking credit consumption across AI models. Located in Account Settings, the tool provides usage metrics for consumer and enterprise accounts. It displays total credits, active member counts, and a breakdown of consumption by model, including Claude and GPT variants, to provide visibility into agentic workflow costs.
PerplexityJul 10Perplexity added OpenAI’s GPT-5.6 Terra and Sol models to its answer engine and Perplexity Computer platform. The integration provides access to OpenAI’s latest flagship and balanced model variants directly within the interface.
PerplexityJul 9Perplexity released a research preview of a new orchestrator model for Perplexity Computer, adapted from GLM 5.2. The model features an advisor tool that natively escalates to stronger models when necessary. It delivers near-frontier performance at 0.344x the cost of Opus, averaging roughly half the cost of Opus across benchmarks while running on Nvidia B200 GPUs.
PerplexityJun 24Perplexity launched Computer for Counsel, a specialized version of its agentic platform designed to automate legal administrative tasks. The system connects to research databases, document tools, and matter-management systems like Midpage, LegalZoom, Docusign, and NetDocuments to pull citable sources for briefs and memos. It is available now for all Perplexity Pro and Max subscribers.
PerplexityJun 18Perplexity launched Brain, a self-improving memory system for its Computer agent platform. Brain builds a context graph of past work to refine future task execution, improving answer correctness by 25% and recall by 16% while reducing task costs by 13%. The system is available now as a research preview for all Perplexity Max subscribers.
PerplexityJun 13Perplexity integrated Deep Research as a native skill within its Computer agent platform. The system uses a Search as Code architecture to run thousands of parallel retrieval steps, allowing it to synthesize reports, decks, and dashboards from primary sources and user files. This update is available now to all Pro and Max subscribers.
PerplexityJun 13Perplexity integrated Anthropic's Claude Fable 5 as an orchestrator model within its Computer platform. Designed for long, complex tasks, the model is available only to Pro and Max subscribers, selectable from Computer's orchestrator dropdown alongside GPT-5.5 and other Claude models to manage autonomous, multi-step agentic workflows.
PerplexityJun 9Perplexity, in collaboration with Harvard Business School, published new research on its Computer autonomous agent. The study found that workers using Computer completed tasks in 87% less time and at 94% lower cost than using Search alone, demonstrating how agents expand the scope and efficiency of knowledge work.
PerplexityJun 5Perplexity has made NVIDIA's Nemotron 3 Ultra model available to its Pro and Max subscribers on its platform and Perplexity Computer. This open model is designed for long-running AI agents, offering faster task completion and reduced costs for complex, multi-step workflows.
PerplexityJun 4Perplexity is launching the Main Street AI Accelerator in collaboration with the U.S. Small Business Administration. The program commits $25 million in Perplexity Computer credits to support up to 100,000 eligible small businesses.
PerplexityJun 4Perplexity is bringing Personal Computer to Windows, an AI agent that will run locally to manage files and applications. The system will allow users to automate workflows across the Microsoft 365 stack and local folders using an orchestration layer of over 20 frontier models.
PerplexityJun 2Perplexity has added native support for Apple Health and Function Health data within its answer engine. Users can now query personal biomarkers and activity metrics to receive health insights grounded in their own biological data. This update transforms the platform into a personalized health assistant capable of tracking long-term medical trends.
PerplexityJun 2Perplexity is launching a hybrid agentic inference system that automatically routes tasks between on-device models and cloud-based frontier models. The update allows sensitive data to remain local while utilizing server-side compute for complex reasoning.
PerplexityJun 1Perplexity released Search as Code, a new architecture that lets AI agents orchestrate search primitives through generated Python code. By moving from serial tool-calling to parallel execution, agents can now perform hundreds of retrieval operations in a single turn to improve research accuracy.
PerplexityMay 28Perplexity launched native integrations for Word, Excel, PowerPoint, and Outlook, embedding its agentic orchestration layer directly into the Microsoft 365 side panel. The update eliminates manual copy-pasting by allowing users to generate research-backed documents, financial models, and slide decks where they already work.
PerplexityMay 28Perplexity integrated Anthropic's latest Claude Opus 4.8 model into its answer engine and agentic Computer platform for top-tier subscribers. This update ensures power users have immediate access to the newest frontier reasoning capabilities within a unified research and execution environment.
PerplexityMay 27Perplexity open-sourced a rebuilt Unigram tokenizer that reduces CPU utilization by five to six times compared to standard implementations. While GPU inference often gets the focus, this update targets the hidden bottleneck of CPU-side tokenization for fast models like rerankers.
PerplexityMay 22Perplexity open-sourced Bumblebee, a read-only security scanner for macOS and Linux designed to identify supply-chain vulnerabilities on developer endpoints. The tool specifically targets the emerging attack surface of AI agent configurations and editor extensions that traditional scanners often miss.
PerplexityMay 20Perplexity launched a new context compression system that surgically extracts query-relevant text from web pages before passing it to its answer models. By culling ads, navigation, and metadata, the system reduces input tokens by up to 70 percent while increasing the density of vital information.
PerplexityMay 19Perplexity showcased a case study where its Computer agent automated project coordination for fintech firm Rho, reducing weekly syncs from 2.5 hours to 15 minutes. By autonomously scanning fragmented tool stacks to flag inconsistencies, the agent shifts from a reactive search engine to a proactive digital worker.
PerplexityMay 15Perplexity launched a Snowflake connector for its Computer agent, allowing users to query live data warehouses using natural language. The integration bypasses traditional SQL bottlenecks by automatically generating a semantic data map to translate business questions into accurate queries. This shifts the platform from a web-search tool into a proactive internal data science team.
PerplexityMay 14Perplexity detailed the security architecture for its autonomous agent, using hardware-level microVMs to isolate every task. This shift from software to hardware isolation aims to prevent agents from leaking credentials or being hijacked by malicious web content.
PerplexityMay 12Perplexity published research showing that NVIDIA's GB200 Blackwell architecture nearly halves communication latency for large Mixture-of-Experts models compared to the previous generation. The findings suggest that Blackwell is a primary platform for reducing the cost and latency of serving frontier-scale AI search.
PerplexityMay 8Perplexity published its internal guide for developing Agent Skills, a modular format used to package domain expertise and procedural knowledge for autonomous systems. The manual argues that effective skill design requires reversing traditional software principles, prioritizing specific failure cases and progressive context loading over simple code.
PerplexityMay 7Perplexity released a new Mac application that brings its Personal Computer agentic capabilities to all users. The system allows the AI to execute multi-step tasks across local files, native desktop apps, and the web simultaneously. By enabling cross-device control from iPhone to Mac, the update shifts the platform from a reactive search engine to a proactive digital worker.
PerplexityMay 7Perplexity launched Finance Search in its Agent API, allowing developers to retrieve licensed financial datasets and real-time market data through a single tool call. By consolidating fragmented data sources into a unified schema, the tool enables agents to perform complex valuation lookups with verifiable citations. This shifts financial agent development from manual data integration to a standardized retrieval layer that prioritizes factual grounding.
PerplexityMay 7Perplexity developed a custom inference engine called ROSE and a domain-specific language to build specialized GPU kernels for NVIDIA hardware. By moving down the stack, the company can achieve peak performance on Blackwell chips and reduce latency for massive trillion-parameter models.
PerplexityMay 5Perplexity launched a specialized version of its agentic platform for finance teams, integrating licensed data from providers like Morningstar and PitchBook. The system automates 35 repetitive analyst workflows while providing direct traceability to primary sources like SEC filings and earnings transcripts. This shifts the platform from a general search tool into a verticalized agent capable of high-stakes institutional research.
PerplexityMay 5Perplexity launched Premium Health Sources, integrating professional medical journals like NEJM and BMJ into its answer engine and Computer agent. By grounding responses in peer-reviewed clinical data rather than general web content, the platform aims to provide the same evidence-based information used by physicians and researchers.
PerplexityMay 4Perplexity launched its Computer agent for Microsoft Teams, allowing users to run research, data analysis, and document generation directly within the chat interface. By embedding its multi-model orchestration layer into the primary enterprise communication hub, Perplexity is positioning itself as a modular alternative to native workplace assistants.
Frequently asked questions