Perplexity 将 Model Council 扩展至 Computer 平台，最多可选 8 个模型

ai and ml

Up to 8 AI models running in the cloud weighing in on ambiguous business issues? Sounds affordable

Hold on tight to your tokens: AI search firm Perplexity on Tuesday expanded its five-month-old Model Council to its cloud-based Computer platform, giving users more ways to configure multi-model queries.

The idea behind the Model Council, Perplexity said in Tuesday’s release, is to “allow users to build a board of top models for a single issue.” Models assigned to a council and given a prompt will tackle the problem independently, resulting in a synthesis report that explains where their conclusions agree and diverge.

According to Perplexity comms chief Jesse Dwyer, Model Council functions as a skill that tells an orchestrator model to spin up parallel subagents, each acting as a full agent harness and tackling the problem independently.

“A wait barrier holds until every perspective finishes, then the model the user selected acts as the ‘chair,’” Dwyer told us in an email. “It reads all the perspective outputs and synthesizes them, explicitly surfacing where the models agree and where they diverge.”

When originally launched back in February, the Model Council was restricted to three models available on Perplexity with no user choice. Ask a query, get an answer put together by a synthesizer model from whichever models aren’t on their smoke break - that was that.

With the addition of Model Council to Computer, users can now choose from two to eight models for their AI advisory board, with options ranging from models supplied by frontier labs such as OpenAI, Anthropic, and Google to open-weight models including GLM and Kimi. Analysis depth can also be customized, and while a synthesizer model still puts together the report, Computer can also “turn that synthesis into outputs like reports, board decks, and other work-ready assets,” Perplexity said.

Computer, for those unfamiliar with it, is essentially a web interface for a model orchestration layer.

Perplexity describes the overhauled and expanded Model Council as being ideal for ambiguous questions involving “judgment calls, tradeoffs, assessing risks or other situations where there’s no one right answer” because model-to-model dialogue “shows where there’s consensus and where there’s conflict.”

“Consensus gives people more confidence to move forward,” the company said. “Gaps show them where to go deeper.”

Model Council has a number of intended use cases, Perplexity said, like legal questions, financial queries, corporate decision making, business growth modeling, and even engineering. Being added to Computer makes it even more useful, apparently, because users can now ask it for additional details to “sharpen the answer,” and ask pointed follow-up questions to get a stronger result.

Perplexity also encourages users to ask the Council for confidence levels to see what different models think of their collective answer. “Disagreements often come from different starting assumptions, so ask each model to state the assumptions behind its conclusions.”

Of course, all of this comes at a price. Access to Model Council was previously restricted to enterprise tiers (either $34/month per seat for Pro or $271 for Max) and individual Max ($200/month) customers. With the expansion into Computer, the Council will now serve individual Max and Pro ($20/month) users, as well as those on enterprise tiers.

MORE CONTEXT

Search-capable AI agents may cheat on benchmark tests; Perplexity offers training wheels for building AI agents; Perplexity rips another page from the Google playbook with its own browser, Comet; Guide for the perplexed – Google is no longer the best search engine

But that’s not the only cost associated with the reformed Model Council, because Computer relies on usage-based billing. Pro users get an initial pool of 4,000 bonus credits, while Max customers get 35k and a renewing 10k a month. Enterprise Pro customers get 8,000 bonus credits and 500 a month per seat, while Enterprise Max customers start with a 45k bonus pool and get 15k per seat per month.

According to Perplexity, 100 Computer credits are equal to $1. A "light" task, like summarizing a document and pulling out key points, can eat between 100 and 350 credits. A "heavy" task, which Perplexity describes as creating work-ready assets like pitch decks, can eat as many as 2,275 credits. Because Computer credits function at the product level rather than mapping directly to raw AI token usage, it's not clear how many credits a Model Council session could burn, and Perplexity didn't get specific.

In other words, if you want the Council to weigh in on one of those aforementioned ambiguous questions, it could cost you. AI tokens can get quite expensive, after all.

“Query complexity will affect token usage more than the specific model selection,” Dwyer explained. “Simple queries in Computer can cost less than a penny while complex tasks run higher.” ®

Fire up that iPhone 1 emulation and party like it's 2007

Browser-based fun to remember a time before Apple's annual cash grab became a thing

New motherboards, old RAM, plenty of life left in yesterday's PCs

DDR4 makes a comeback as component costs rise, while Linux offers an escape from forced upgrades

The VMware exit is a protection upgrade

PARTNER CONTENT: VergeIO says an exit is also an opportunity

Tiny11 puts Windows 11 on a diet without the PowerShell workout

New desktop app simplifies stripping out bundled software, with heavier cuts carrying security and support risks

It won't kill you, or humanity, to get off the AI upgrade treadmill

New models arrive so fast, we already have more new tech than it's possible to understand in a career

Cisco, Nvidia team up to build fatter, faster pluggable

If you thought 1.6 Tbps was fast, just wait, Cisco's USFP targets 12.8 Tbps in a single optical pluggable

TOP STORIES

Anthropic asks users to stop being mean to Claude; '123456' password linked to Denmark data breach affecting 8.8M people; Microsoft and Anthropic play invoice tennis with startup's $17,600 Claude bill; Starlink's plan to avoid orbital near-misses: Ephemeris sharing; Shai-Hulud worm makes jump to AI infrastructure with Tensorlake compromise; Dutch tax office ditches Microsoft 365 cloud for on-premises alternative

Cisco, Nvidia team up to build fatter, faster pluggable If you thought 1.6 Tbps was fast, just wait, Cisco's USFP targets 12.8 Tbps in a single optical pluggable; Claude submitted fake murder tip to Philly police during summertime testing Anthropic took more than two months to alert cops; AI systems are fully capable of carrying out nightmare attacks against infrastructure and nobody's ready Meanwhile, AI agents move robotic arms, 'make OT device operator screens lie'; Nvidia found $1B under the couch to help secure American scientific computing dominance Commitment comes as GPUzilla prepares to fortify Uncle Sam's arsenal with at least seven AI-optimized supers; TSMC taps GlobalFoundries to bolster US silicon interposer production in $2B deal Advanced packaging tech is essential to the domestic production of high-performance semiconductors used in AI datacenters

Infosec

Security Russians are posing as Signal support to launch phishing attacks PLUS: US takes down Iranian propaganda sites; Marketing company asks 'Why Do We Have Your Information?' And more!; Security Microsoft patches failed to fix on-prem SharePoint, which is now under zero-day attack PLUS: China upgrades smartphone surveillance tools; Ring eases anti-snooping stance; and more; Black Hat and DEF CON DEF CON Franklin project enlists hackers to harden critical infrastructure Voting village reports have been so successful, says Jeff Moss, that the whole of DEF CON will now be included; Security EQT buys majority share in Swiss cybersecurity biz Acronis Went at equivalent of $3.5B+ valuation for entire firm, though portion sold not specified; Malware Month Ten years since the first corp ransomware, Mikko Hyppönen sees no end in sight On the plus side, infosec's a good bet for a long, stable career

FOSS

NetBSD may have found a new role: The AI-free FOSS OS Increasing awareness and adoption, followed by increasing donations; OpenAI dots inspire open source imitators amid technical difficulties Persistent agents appeal to AI companies, but demand isn't obvious; Ubuntu 'Stonking Stingray' beta swims out: Deeper Rust 'oxidization' plus LLM speech-to-text GNOME 51 is here, and Ubuntu interim releases are getting interesting again; Revive Raskin's 3 laws of software with humane work that considers a dev's duty of care All this, and angry NeoVim users; Firefox 157: Could bold new look be just what Mozilla needs?* Yet another new version of Firefox, with a significant reskin; KDE turns 30 and someone's brought an AI-native desktop proposal Akademy talk imagines Plasma assembling itself around a personal model of each user