Skip to content
AI EmployeeSuper-AgentsAgent-to-AgentBenchmarksPricingBlog

Perplexity Multiagent: Mixed-Model Teams, Benchmarked

At a glanceQuick answers
What is Perplexity Multiagent?
A new effort level in Perplexity Computer and teammates. A lead agent splits a task across agents on different frontier models and combines their findings into one answer.
Does it score higher?
In Perplexity’s own runs, every mixed team beat four copies of its lead model on Terminal-Bench 4.1 and on WANDR.
What does it cost?
Perplexity calls it credit-intensive and has not published a price. Its top teams cost less per run than four copies of the lead model.
Editorial illustration on a near-white ground titled Perplexity Multiagent. A large teal square, the lead agent, is linked to three smaller squares in amber, coral and violet that pass arrows between them, beside two large figures: 0.727, top mixed team on Terminal-Bench 4.1, and +0.125, the best gain of a mixed team over four copies.
Fig 0Perplexity's Multiagent results in two figures. Made by CellCog's image agent, running GPT Image 2.6.

Perplexity has added Multiagent, a new effort level in Computer and teammates. A lead agent splits a task across a team of agents that run on different frontier models, the agents share findings as they work, and the lead combines everything into one answer. Perplexity announced it on X at 16:28 UTC on November 17, 2026, with benchmark charts in the same thread.

The numbers carry the story. In Perplexity’s own runs, four copies of one model were not reliably better than one, and for two models they scored lower. A team that mixed models beat four copies of its lead model every time. This page records what Perplexity published and what it left out.

On this page · 7 sectionsOpen
  1. What Perplexity launched
  2. Terminal-Bench 4.1: copies against a mixed team
  3. WANDR: score, cost and time
  4. What Perplexity did not publish
  5. How it fits the debate on teams of agents
  6. Where CellCog sits
  7. Sources
Key points5 · 7 min full read
  1. A large square linked to three smaller squares in different colours.
    Perplexity launched Multiagent on November 17, 2026: a new effort level in Computer and teammates where a lead agent runs a team of agents on different frontier models.
  2. A small bar chart with the tallest bar marked.
    On Terminal-Bench 4.1, every mixed team beat four copies of its lead model, by 0.008 to 0.125, in Perplexity’s own runs.
  3. Four identical squares beside a downward arrow.
    Four copies of Claude Opus 5.5 or GPT-6.1 Sol scored lower than one agent of the same model.
  4. A stack of coins beside a stopwatch.
    The top team, led by Gemini 4 Carbon, scored 0.727 at $10.31 a run. Perplexity says four Gemini copies cost about $13.
  5. A gauge with its needle near empty.
    Perplexity calls Multiagent credit-intensive and says to use it only for the most important tasks. It has not published a price or a plan list in the thread.

§ 01What Perplexity launched

  • The mode. “We’re rolling out Multiagent, a new effort level in Computer and teammates.”
  • How it works. “A lead agent splits tasks across a team of agents using different frontier models.” Then: “As they work, agents share findings with one another, with the lead agent combining everything into one answer.”
  • Who it is for. “Multiagent is optimized for performance.” And: “Multiagent is best for forecasting, high-density problem solving, and consequential decision making.”
  • The cost. “It is credit-intensive. Use it only for your most important tasks.” The thread gives no credit figure.
  • Where it runs. In Perplexity Computer and in teammates, the agents Perplexity launched on November 5. Our teammates record covers that launch.

The launch post had 3,419 views when we read it through the X API shortly after it went up.

Animated isometric scene titled One lead, a mixed team. A task card arrives at a teal lead block, which sends three pieces down dashed lines to three different agent blocks, white, grey and black, on a platform labelled Different models. Findings hop between the agents and rise back to the lead, and a card marked One answer comes out. The last frame reads Many models, one answer.
Fig 1How Multiagent works: a lead agent splits one task across a team on different models.

§ 02Terminal-Bench 4.1: copies against a mixed team

Perplexity says “we ran five frontier models alone, as two and four copies, and leading a mixed-model team of four”. Its finding: “each mixed-model team outperformed both the lead model and multiple copies”. The table gives the values printed on its charts. The gain column in the chart below is Perplexity’s own printed figure for each mixed team over four copies of its lead.

Lead model Alone Four copies Mixed team Mixed team cost per run
Gemini 4 Carbon 0.644 0.652 0.727 $10.31
Claude Fable 5.2 0.568 0.614 0.629 $25.23
Claude Opus 5.5 0.500 0.496 0.621 $4.53
GPT-6 Astra 0.508 0.561 0.568 $11.27
GPT-6.1 Sol 0.470 0.462 0.545 $3.18
Table 1Terminal-Bench 4.1 scores and cost per run, as printed on Perplexity’s charts, November 17, 2026
Gain of each mixed team over four copies of its lead model, Terminal-Bench 4.1Bar chart of the gain each mixed team made over four copies of its lead model on Terminal-Bench 4.1: Claude Opus 5.5 highlighted at 0.125, then GPT-6.1 Sol 0.083, Gemini 4 Carbon 0.076, Claude Fable 5.2 0.015 and GPT-6 Astra 0.008Claude Opus 5.50.125GPT-6.1 Sol0.083Gemini 4 Carbon0.076Claude Fable 5.20.015GPT-6 Astra0.008Gain of each mixed team over four copies of its lead model, Terminal-Bench 4.1Bar chart of the gain each mixed team made over four copies of its lead model on Terminal-Bench 4.1: Claude Opus 5.5 highlighted at 0.125, then GPT-6.1 Sol 0.083, Gemini 4 Carbon 0.076, Claude Fable 5.2 0.015 and GPT-6 Astra 0.008Claude Opus 5.50.125GPT-6.1 Sol0.083Gemini 4 Carbon0.076Claude Fable 5.20.015GPT-6 Astra0.008
Fig 2Gain of each mixed team over four copies of its lead model, Terminal-Bench 4.1
  • Copies did not always help. Four copies of Claude Opus 5.5 scored 0.496 against 0.500 alone. Four copies of GPT-6.1 Sol scored 0.462 against 0.470.
  • Mixing did. Every mixed team beat four copies of its lead, by 0.008 for GPT-6 Astra up to 0.125 for Claude Opus 5.5.
  • The top score. The team led by Gemini 4 Carbon scored 0.727 at $10.31 a run. Perplexity says it “The top mixed team on each also cost less than four copies of its lead”: $10.31 against about $13.
  • The cheap end. The team led by GPT-6.1 Sol cost $3.18 a run and scored 0.545, above every result for Sol alone or in copies.

§ 03WANDR: score, cost and time

Perplexity repeated the test on a second benchmark, WANDR. Gemini 4 Carbon does not appear in its WANDR charts. “Every mixed team beat four copies of its lead model” on both benchmarks, the thread says: by 0.015 to 0.064 on WANDR.

Lead model Mixed team Gain over four copies Mixed team cost per run Median minutes, one agent to four copies
Claude Fable 5.2 0.868 +0.015 $54.50 64.3 to 48.2
GPT-6 Astra 0.837 +0.035 $43.77 37.8 to 19.0
GPT-6.1 Sol 0.809 +0.042 $10.47 33.3 to 17.3
Claude Opus 5.5 0.801 +0.064 $11.52 14.7 to 15.9
Table 2WANDR mixed-team results and minutes per run, as printed on Perplexity’s charts, November 17, 2026
Mixed-team scores on WANDR by lead model, as printed by PerplexityBar chart of mixed-team scores on WANDR: Claude Fable 5.2 highlighted at 0.868, GPT-6 Astra 0.837, GPT-6.1 Sol 0.809 and Claude Opus 5.5 0.801Claude Fable 5.20.868GPT-6 Astra0.837GPT-6.1 Sol0.809Claude Opus 5.50.801Mixed-team scores on WANDR by lead model, as printed by PerplexityBar chart of mixed-team scores on WANDR: Claude Fable 5.2 highlighted at 0.868, GPT-6 Astra 0.837, GPT-6.1 Sol 0.809 and Claude Opus 5.5 0.801Claude Fable 5.20.868GPT-6 Astra0.837GPT-6.1 Sol0.809Claude Opus 5.50.801
Fig 3Mixed-team scores on WANDR by lead model, as printed by Perplexity
  • The top team. Led by Claude Fable 5.2, at 0.868 and $54.50 a run. Perplexity puts four Fable copies at $66.41.
  • Time. “Every team of four scored higher than its model alone, and three of the four finished sooner.” The minutes chart compares one agent with four copies of it, not with a mixed team. Four copies of GPT-6 Astra took 19.0 minutes against 37.8 for one agent.
  • The exception. Claude Opus 5.5 was the one that did not finish sooner: 14.7 minutes alone, 15.9 as four copies. It was also the fastest of the four in both cases.

§ 04What Perplexity did not publish

  • The rest of each team. The charts name only the lead model. The thread does not say which models joined it.
  • Run counts. No number of runs per score and no spread between runs.
  • WANDR itself. The thread does not describe the benchmark, and we did not look it up for this page.
  • Price and plans. No credit figure and no list of plans that get the mode.
  • An outside check. These are Perplexity’s own runs on its own product. We have not reproduced any of them.

§ 05How it fits the debate on teams of agents

Anthropic’s October guide, in our record, advised folding subagents into the main loop for most tasks. Perplexity’s copies result sits close to that: more of the same model was not reliably better. Its mixed-team result is a different claim, that models from different labs working together beat any one of them in copies. That is our reading of the two documents, not a statement from either company.

Multi-agent products are ranked in our list of multi-agent platforms, which we keep current.

§ 06Where CellCog sits

Our conflict, declared: we build CellCog, a platform where a business hires AI employees, and those employees run on Anthropic’s Claude models. CellCog’s Agent Team mode puts two agents with different jobs on one problem, one directing and one doing the work, both on Claude Opus 5.5. It does not mix models from different labs. We have not run Multiagent or Perplexity’s benchmarks.

§ 07Sources

Frequently asked5 questions

Q1What is Perplexity Multiagent?

An effort level in Perplexity Computer and teammates, announced on November 17, 2026. A lead agent splits a task across a team of agents that use different frontier models, they share findings while they work, and the lead combines everything into one answer.

Q2Which model led the best team?

On Terminal-Bench 4.1, the mixed team led by Gemini 4 Carbon scored highest, at 0.727. On WANDR, the team led by Claude Fable 5.2 scored highest, at 0.868.

Q3Do more copies of the same model help?

Not reliably, in Perplexity’s numbers. On Terminal-Bench 4.1, four copies of Claude Opus 5.5 scored 0.496 against 0.500 for one, and four copies of GPT-6.1 Sol scored 0.462 against 0.470.

Q4How much does Multiagent cost?

Perplexity has not published a price in the launch thread. It says the mode is credit-intensive and should be used only for the most important tasks.

Q5Are these independent benchmark results?

No. They are Perplexity’s own runs of its own product. The thread does not give run counts, variance, or which models joined each lead.

Published 17 November 2026 All Multi-agent & AI organizations →