This media is not supported in your browser
VIEW IN TELEGRAM
Portable Computer is now available on Windows PCs with NVIDIA RTX GPUs.
Run the harness, agents, and models locally on your PC.
Work with local files and connected apps without sending tasks to the cloud. Use frontier cloud models when needed.
Run the harness, agents, and models locally on your PC.
Work with local files and connected apps without sending tasks to the cloud. Use frontier cloud models when needed.
🔥4
Perplexity Computer will now come pre-installed on HP ZBook Ultra G3a.
Use Computer to run complex multi-step work from a simple interface. Computer agents are grounded in accurate deep research and connected to hundreds of tools, now including Autodesk.
On supported HP PCs, users can use Portable Computer, our local-first version of Computer, to run tasks entirely on-device with a local model.
Portable Computer keeps private data on your machine without using cloud credits. Read more: https://pplx.ai/hp-windows-tg
Use Computer to run complex multi-step work from a simple interface. Computer agents are grounded in accurate deep research and connected to hundreds of tools, now including Autodesk.
On supported HP PCs, users can use Portable Computer, our local-first version of Computer, to run tasks entirely on-device with a local model.
Portable Computer keeps private data on your machine without using cloud credits. Read more: https://pplx.ai/hp-windows-tg
👍4🙏1
This media is not supported in your browser
VIEW IN TELEGRAM
A new Perplexity Search SDK cookbook is live.
The recipe fans out focused searches, filters results to official docs, extracts relevant passages, and writes a source-linked brief for your coding agent.
Get started: https://pplx.ai/search-sdk-tg
The recipe fans out focused searches, filters results to official docs, extracts relevant passages, and writes a source-linked brief for your coding agent.
Get started: https://pplx.ai/search-sdk-tg
👍3
This media is not supported in your browser
VIEW IN TELEGRAM
We’re rolling out effort controls in Computer’s model selector.
Effort presets combine the orchestrator model and reasoning depth to control how deeply and efficiently Computer works through a task.
Available now on web. Coming to mobile and desktop soon.
Effort presets combine the orchestrator model and reasoning depth to control how deeply and efficiently Computer works through a task.
Available now on web. Coming to mobile and desktop soon.
🔥2
This media is not supported in your browser
VIEW IN TELEGRAM
Perplexity Computer can now create videos using MiniMax H3 and ByteDance Seedance 2.5.
Ask Computer for a campaign clip, product demo, or social asset, and it produces finished video next to copy and creative in same thread.
Available now for Pro and Max subscribers.
Ask Computer for a campaign clip, product demo, or social asset, and it produces finished video next to copy and creative in same thread.
Available now for Pro and Max subscribers.
👏3❤1
We've shipped 15 major improvements to Computer in September so far.
Effort controls - pick Light, Standard, High, or Ultra and Computer chooses the model and reasoning depth. Click Custom under the slider to select them yourself. Live on web, mobile and desktop soon.
GPT-6 Astra - now available for Computer tasks and supporting agents for eligible Pro and Max subscribers.
Portable Computer on Windows and Linux - run Computer entirely on your own PC, no cloud dependency. Local work uses no credits.
Hybrid compute on Mac - split a task between the cloud and a private on-device model. Sensitive steps stay on your Mac, only de-identified results go to the cloud.
Skills marketplace - browse and install reusable Computer skills, and share them across your Enterprise org. Find it under Customize > Skills.
Side chat - ask a quick question about a running task in a read-only side chat with /ask, /side, or /btw, without interrupting the main task.
Also new: Cmd+K history search, Computer turns for eligible US non-paying accounts, voice noise isolation on iPhone and Mac, Computer for Builders, Microsoft 365 in Ask, Android setup checklist, Omnisend, Higgsfield, and Evernote connectors, and Enterprise usage analytics.
Read about all of our recent launches in the September changelog: https://pplx.ai/sept-changelog-tg
Effort controls - pick Light, Standard, High, or Ultra and Computer chooses the model and reasoning depth. Click Custom under the slider to select them yourself. Live on web, mobile and desktop soon.
GPT-6 Astra - now available for Computer tasks and supporting agents for eligible Pro and Max subscribers.
Portable Computer on Windows and Linux - run Computer entirely on your own PC, no cloud dependency. Local work uses no credits.
Hybrid compute on Mac - split a task between the cloud and a private on-device model. Sensitive steps stay on your Mac, only de-identified results go to the cloud.
Skills marketplace - browse and install reusable Computer skills, and share them across your Enterprise org. Find it under Customize > Skills.
Side chat - ask a quick question about a running task in a read-only side chat with /ask, /side, or /btw, without interrupting the main task.
Also new: Cmd+K history search, Computer turns for eligible US non-paying accounts, voice noise isolation on iPhone and Mac, Computer for Builders, Microsoft 365 in Ask, Android setup checklist, Omnisend, Higgsfield, and Evernote connectors, and Enterprise usage analytics.
Read about all of our recent launches in the September changelog: https://pplx.ai/sept-changelog-tg
Perplexity AI
Effort Mode, GPT-6 Astra & Skills Marketplace | Perplexity
Choose Computer’s effort level, use GPT-6 Astra, discover reusable skills, and run Hybrid compute on Mac, with new connectors and tools for teams.
This media is not supported in your browser
VIEW IN TELEGRAM
Portable Computer for Windows is now available on AMD Ryzen AI Max Series processors.
It makes it easy to run local AI agents that work with your connected apps and local files. Kick off tasks or schedule recurring work that runs entirely on your device.
It makes it easy to run local AI agents that work with your connected apps and local files. Kick off tasks or schedule recurring work that runs entirely on your device.
Introducing Fast Search in the Perplexity Search API.
Fast Search runs on Photon, our new Rust-based retrieval and ranking service that we built with a small team of engineers and hundreds of agents.
It returns 95% of search results in 230 ms or less.
How we built Photon:
https://perplexity.ai/hub/blog/photon
Get started with Fast Search:
https://pplx.ai/fast-search-api-tg
Fast Search runs on Photon, our new Rust-based retrieval and ranking service that we built with a small team of engineers and hundreds of agents.
It returns 95% of search results in 230 ms or less.
How we built Photon:
https://perplexity.ai/hub/blog/photon
Get started with Fast Search:
https://pplx.ai/fast-search-api-tg
❤4
We’re partnering with Nvidia and 100+ industry partners to build infrastructure that contains rogue AI agents.
In this research, we gave 9 AI models root access inside SPACE and told them to break out.
Across 108 runs, none breached the VM boundary.
https://pplx.ai/escaping-space-tg
In this research, we gave 9 AI models root access inside SPACE and told them to break out.
Across 108 runs, none breached the VM boundary.
https://pplx.ai/escaping-space-tg
❤2
This media is not supported in your browser
VIEW IN TELEGRAM
You can now build custom reusable agents in the Perplexity Agent API with Profiles, Skills, and managed connectors.
Configure an agent once in the API Portal and reuse it across applications and workflows.
A Profile saves the agent's model, instructions, tools, Skills, and managed connectors under one versioned ID.
Read more about Profiles:
https://pplx.ai/api-agents-tg
Configure an agent once in the API Portal and reuse it across applications and workflows.
A Profile saves the agent's model, instructions, tools, Skills, and managed connectors under one versioned ID.
Read more about Profiles:
https://pplx.ai/api-agents-tg
🔥2👏1
Agent governance is an engineering problem.
We’ve built safeguards into Perplexity’s infrastructure, harnesses, and tools, and put them to work across our products.
Today we're sharing how we engineer safer agents:
https://pplx.ai/safer-agents-tg
We’ve built safeguards into Perplexity’s infrastructure, harnesses, and tools, and put them to work across our products.
Today we're sharing how we engineer safer agents:
https://pplx.ai/safer-agents-tg
Perplexity AI
How we engineer safer agents
Layered safeguards, independent monitoring, and human oversight across Perplexity’s cloud, browser, and endpoint agents.