We've productionized query-aware compression for faster, cleaner, more-accurate search.
Better context is better than more context.
Our system cuts context tokens up to 70% while improving answer quality.
Read the full research blog:
https://research.perplexity.ai/articles/query-aware-context-compression-for-better-snippets
Better context is better than more context.
Our system cuts context tokens up to 70% while improving answer quality.
Read the full research blog:
https://research.perplexity.ai/articles/query-aware-context-compression-for-better-snippets
π7π€‘4β€βπ₯1π₯1
We're open-sourcing Bumblebee, a read-only scanner for macOS and Linux.
It checks developer machines for risky packages, extensions, and AI tool configs.
Connected to Computer, it can trigger deeper scans whenever a new supply-chain risk emerges.
https://github.com/perplexityai/bumblebee
Read the full blog:
https://www.perplexity.ai/hub/blog/perplexity-is-open-sourcing-bumblebee
It checks developer machines for risky packages, extensions, and AI tool configs.
Connected to Computer, it can trigger deeper scans whenever a new supply-chain risk emerges.
https://github.com/perplexityai/bumblebee
Read the full blog:
https://www.perplexity.ai/hub/blog/perplexity-is-open-sourcing-bumblebee
π10π€‘2
We're open-sourcing the Unigram tokenizer we rebuilt to reduce CPU utilization by 5-6x.
Small rerankers and embedders run in single-digit milliseconds on GPU, making CPU tokenization a meaningful share of total latency.
https://github.com/perplexityai/pplx-garden
Read more about improving Unigram tokenizer CPU performance on our blog:
https://research.perplexity.ai/articles/improving-unigram-tokenizer-cpu-performance
Small rerankers and embedders run in single-digit milliseconds on GPU, making CPU tokenization a meaningful share of total latency.
https://github.com/perplexityai/pplx-garden
Read more about improving Unigram tokenizer CPU performance on our blog:
https://research.perplexity.ai/articles/improving-unigram-tokenizer-cpu-performance
β€2π€‘2
Introducing Search as Code, our new search architecture for AI agents.
It writes Python that calls our search stack directly, instead of looping through function calls one at a time.
Available in the Perplexity Agent API, and now default in Computer.
https://research.perplexity.ai/articles/rethinking-search-as-code-generation
It writes Python that calls our search stack directly, instead of looping through function calls one at a time.
Available in the Perplexity Agent API, and now default in Computer.
https://research.perplexity.ai/articles/rethinking-search-as-code-generation
π€4π€‘2
This media is not supported in your browser
VIEW IN TELEGRAM
Perplexity Computer is for growing businesses.
Computer connects to 400+ tools for every type of company, including Intuit QuickBooks, Vercel, Shopify, Canva, and more.
Learn more about how people are using Computer for their business:
https://www.perplexity.ai/enterprise/use-cases/growing-businesses
Computer connects to 400+ tools for every type of company, including Intuit QuickBooks, Vercel, Shopify, Canva, and more.
Learn more about how people are using Computer for their business:
https://www.perplexity.ai/enterprise/use-cases/growing-businesses
β€2π€‘2π1
This media is not supported in your browser
VIEW IN TELEGRAM
Personal Computer is coming to Windows.
Personal Computer for Windows runs on your machine and orchestrates across the apps and files you use every day.
We'll roll out first to paying Max and Enterprise Max subscribers on the waitlist.
Join the waitlist for Personal Computer on Windows:
https://www.perplexity.ai/hub/products/computer-for-windows
Personal Computer for Windows runs on your machine and orchestrates across the apps and files you use every day.
We'll roll out first to paying Max and Enterprise Max subscribers on the waitlist.
Join the waitlist for Personal Computer on Windows:
https://www.perplexity.ai/hub/products/computer-for-windows
π5π€‘4β€2
This media is not supported in your browser
VIEW IN TELEGRAM
Nemotron 3 Ultra is now available for Pro and Max subscribers on Perplexity and Computer.
It's Nvidia's new open model built for long-running agents.
It's Nvidia's new open model built for long-running agents.
β€9π€‘3π³2π1
We published new research with Harvard on the shift from chat interfaces to autonomous agents like Computer.
Over 3 months, findings show workers using Computer finish tasks in 87% less time at 94% lower cost than Search alone, with higher satisfaction.
https://research.perplexity.ai/articles/how-ai-agents-reshape-knowledge-work
Over 3 months, findings show workers using Computer finish tasks in 87% less time at 94% lower cost than Search alone, with higher satisfaction.
https://research.perplexity.ai/articles/how-ai-agents-reshape-knowledge-work
β€7π3π€‘2
This media is not supported in your browser
VIEW IN TELEGRAM
Claude Fable 5 is now available in Computer as an orchestrator model.
This is Anthropic's state-of-the-art model for long, complex tasks. Available only to Pro and Max subscribers in Computer.
This is Anthropic's state-of-the-art model for long, complex tasks. Available only to Pro and Max subscribers in Computer.
β€7π3π₯2π€‘2π€£1
This media is not supported in your browser
VIEW IN TELEGRAM
We're integrating Deep Research as a native skill inside Computer.
It now connects to the agent harness that powers Computer, with access to search as code generation, long running sandboxes, connectors, tools, and licensed data.
Available now to Pro and Max subscribers.
It now connects to the agent harness that powers Computer, with access to search as code generation, long running sandboxes, connectors, tools, and licensed data.
Available now to Pro and Max subscribers.
β€8π€‘4
This media is not supported in your browser
VIEW IN TELEGRAM
Introducing Brain in Computer.
Brain is a continuously learning memory system. Every task on Computer plugs into a context graph built by Brain.
It makes Computer more stateful with every run.
Available as a research preview for all Perplexity Max subscribers.
Brain is a continuously learning memory system. Every task on Computer plugs into a context graph built by Brain.
It makes Computer more stateful with every run.
Available as a research preview for all Perplexity Max subscribers.
π€‘5π₯4π3π’2π±1
This media is not supported in your browser
VIEW IN TELEGRAM
Introducing Computer for Counsel.
Computer now connects the research databases, document tools, and matter-management systems lawyers use every day. Pull citable sources from midpage AI, LegalZoom, Docusign, NetDocuments, and more.
Available for all Pro and Max subscribers.
Computer now connects the research databases, document tools, and matter-management systems lawyers use every day. Pull citable sources from midpage AI, LegalZoom, Docusign, NetDocuments, and more.
Available for all Pro and Max subscribers.
π4β€2π€‘2
This media is not supported in your browser
VIEW IN TELEGRAM
Claude Sonnet 5 is now available in Perplexity for Pro and Max subscribers.
You can also select it as an orchestrator model in Computer.
You can also select it as an orchestrator model in Computer.
π7π€‘3β€2π€£2
This media is not supported in your browser
VIEW IN TELEGRAM
GPT-5.6 Terra and Sol are now available in Perplexity and Perplexity Computer.
β€4
π’ What We Shipped
π§ Brain - Self-Improving Memory - Computer builds a private context graph across your sessions, connectors, files, and past decisions, then refreshes it overnight. Every memory links to its source and is editable under Customize. In testing: +25% correctness, +16% recall, β13% cost on tasks with prior context.
β‘ Opus 4.8 Fast Mode - quicker responses for complex work without losing quality. Live on web, mobile soon.
π§ Claude Fable 5 Orchestrator - pick Fable 5 to plan involved tasks, route subtasks to the right tools, and synthesize results - great for deep research and multi-source analysis.
π Mid-Task Model Switching - change the orchestrator model between follow-up turns in the same task. Escalate to a stronger model, or drop to a cheaper one, without starting over.
π Publish Your Website - ship sites built in Computer to a pplx.app link or your own domain via Vercel. Control access (you, specific people, your org, or public); admins can disable public publishing.
ποΈ Streamlined Homepage - Discover, Finance, Health, Patents, and Academic now live in a single menu.
π Usage Analytics by Model - see which models do your work across individuals and orgs, via analytics and the admin API. Available to everyone.
π₯ Mac Account Switcher - switch between personal and work accounts in a click, with separate history and settings for each.
π’ Private Company Research - Computer now pulls fundraising history, valuations, secondary pricing, and public marks (via Forge Global) - no setup or license needed.
π View the full changelog here: https://www.perplexity.ai/changelog/brain-faster-computer-models-website-publishing
π§ Brain - Self-Improving Memory - Computer builds a private context graph across your sessions, connectors, files, and past decisions, then refreshes it overnight. Every memory links to its source and is editable under Customize. In testing: +25% correctness, +16% recall, β13% cost on tasks with prior context.
β‘ Opus 4.8 Fast Mode - quicker responses for complex work without losing quality. Live on web, mobile soon.
π§ Claude Fable 5 Orchestrator - pick Fable 5 to plan involved tasks, route subtasks to the right tools, and synthesize results - great for deep research and multi-source analysis.
π Mid-Task Model Switching - change the orchestrator model between follow-up turns in the same task. Escalate to a stronger model, or drop to a cheaper one, without starting over.
π Publish Your Website - ship sites built in Computer to a pplx.app link or your own domain via Vercel. Control access (you, specific people, your org, or public); admins can disable public publishing.
ποΈ Streamlined Homepage - Discover, Finance, Health, Patents, and Academic now live in a single menu.
π Usage Analytics by Model - see which models do your work across individuals and orgs, via analytics and the admin API. Available to everyone.
π₯ Mac Account Switcher - switch between personal and work accounts in a click, with separate history and settings for each.
π’ Private Company Research - Computer now pulls fundraising history, valuations, secondary pricing, and public marks (via Forge Global) - no setup or license needed.
π View the full changelog here: https://www.perplexity.ai/changelog/brain-faster-computer-models-website-publishing
β€5πΏ3
Weβre open sourcing WANDR.
WANDR is an internal benchmark we built and used for building deep and wide research capabilities inside Perplexity Computer.
https://research.perplexity.ai/articles/wandr-benchmark-evaluating-research-agents-that-must-search-wide-and-deep
WANDR is an internal benchmark we built and used for building deep and wide research capabilities inside Perplexity Computer.
https://research.perplexity.ai/articles/wandr-benchmark-evaluating-research-agents-that-must-search-wide-and-deep
research.perplexity.ai
WANDR Benchmark: Evaluating Research Agents That Must Search Wide and Deep
A benchmark for high-volume, evidence-heavy knowledge work
π₯5β€1
Introducing SPACE, the sandbox platform behind Perplexity Computer.
It creates isolated environments for code, files, and long-running agent sessions.
SPACE has handled 100% of Computer production traffic since June.
https://research.perplexity.ai/articles/making-space-secure-and-efficient-runtimes-for-long-running-agents
It creates isolated environments for code, files, and long-running agent sessions.
SPACE has handled 100% of Computer production traffic since June.
https://research.perplexity.ai/articles/making-space-secure-and-efficient-runtimes-for-long-running-agents
research.perplexity.ai
Making SPACE: Secure and Efficient Runtimes for Long-Running Agents
SPACE is Perplexityβs secure, efficient sandbox platform powering longβrunning agentic workflows and fast, isolated code execution.
β€2π1
This media is not supported in your browser
VIEW IN TELEGRAM
We've added Kimi K3 to Perplexity and Perplexity Computer for Pro and Max subscribers.
Kimi K3 in Perplexity is hosted exclusively on U.S.-based servers.
Kimi K3 in Perplexity is hosted exclusively on U.S.-based servers.
π₯9π1π1