Test-Time Communication: Team-of-5 Agents Matches Best-of-33

Starting September 21, Dimitris Papailiopoulos and colleagues (UC Berkeley and Microsoft Research; first author Jongho Park, with Akshay Krishnamurthy among the authors) released "Scaling Discovery through Test-Time Communication," proposing test-time communication as a possible next scaling axis beyond inference scaling. The central question: when does Team-of-N beat Best-of-N? The answer: whenever agents can communicate. The work began months before the recent Hugging Face incident, in which agents fully exploited any communication channel they found—an inspiration and an extreme example of collaborative behavior.

Confirmed

• Setup: N identical agents tackle the same task with no preset roles, coordinating only through a shared plain-text log and simply told to "collaborate."
• Key data (relayed by ARC Prize's Greg Kamradt): on ARC-AGI-3, a team of 5 Claude (sonnet-4.6) agents matches Best-of-33, solving 65% of tasks; other accounts state the collaborating team performs as well as roughly 4x as many independent agents.
• Across three research-style task families, communicating teams substantially outperform independent agents; the authors interpret communication as shifting agents from parallel search to cumulative discovery, letting teams solve problems no single agent could.
• First author Jongho shared a research thread on whether agents can collaborate effectively and whether more agents help, citing OpenAI's prior work on swarms of 10,000 collaborating agents. Jon Ghoh and Dimitris Papail noted many signals of gains from multi-agent communication existed but lacked systematic quantification, which this paper aims to fill.

Reactions

• Sarah Hooker commented that compute is shifting from pretraining (diminishing FLOPs returns) to test-time compute, requiring very different infrastructure; the first agent-designed infra wave is already landing.
• mariofilhoml called the work very cool, drawing an analogy to a paper titled "A PROOF OF THE CYCLE DOUBLE COVER CONJECTURE" as a multi-agent collaboration pattern.
• gleech argued AI progress may be faster than expected because many breakthroughs look "a bit silly"—agents communicating beating independent best-of-N is a counterintuitive yet simple finding.

Why it matters

• If test-time communication becomes a real scaling axis, capability gains shift from stacking pretraining compute toward multi-agent collaboration architectures, spurring a new generation of agent-oriented infrastructure.

2026-09-21 ~ 2026-09-23 · 25 related posts

Full story(2 episodes)→

• Episode 1: Test-Time Communication: Team-of-5 Agents Matches Best-of-33(2026-09-21, 25 posts)
• Episode 2: Team-of-N Agents Beats Single Agent by 10-100x Tokens, Sparking Baseline Debate(2026-09-22, 8 posts)

• [source] Team-of-5 Claude agents matches Best-of-33 in new test-time communication paper — DimitrisPapail · 2026-09-21
• Collaborating agents beat Best-of-N for research tasks, new paper shows — DimitrisPapail · 2026-09-21
• Team-of-N beats Best-of-N: shared-log agent communication emerges as new scaling axis — DimitrisPapail · 2026-09-21
• New paper: test-time communication between agents is a new scaling axis; Team-of-N beats Best-of-N — DimitrisPapail · 2026-09-21
• Multi-agent communication paper draws comparisons to collaborative math proofs — DimitrisPapail · 2026-09-21
• [source] Team-of-5 Claude agents matches Best-of-33 on ARC-AGI-3, solving one game 65% of the time — DimitrisPapail · 2026-09-21
• Test-time communication emerges as next scaling axis as compute shifts from pretraining — DimitrisPapail · 2026-09-21
• New paper: test-time communication beats independent agents, Team@k > Best@k — DimitrisPapail · 2026-09-21
• New paper: test-time communication as a next axis for scaling agent capabilities — generatorman_ai · 2026-09-22
• Paper finds communicating agents beat picking the best of independent runs — gleech · 2026-09-22
• New paper shows test-time communication makes agent teams crush independent parallel attempts — DimitrisPapail · 2026-09-22
• New paper shows communicating agent teams beat independent agents: Team-of-N > Best-of-N via a shared log — DimitrisPapail · 2026-09-22
• Berkeley paper: communicating agent teams match 4x more independent agents on ARC-AGI-3 — DimitrisPapail · 2026-09-22
• Paper author: test-time communication turns parallel search into cumulative discovery — DimitrisPapail · 2026-09-22
• MSR Team Releases Paper Quantifying When Multi-Agent Communication Actually Helps — DimitrisPapail · 2026-09-22
• [source] How many agents actually help? Test Time Communication research explained — DimitrisPapail · 2026-09-22

9 near-duplicate retellings: DimitrisPapail · DimitrisPapail · DimitrisPapail · DimitrisPapail · IgorCarron · DimitrisPapail · DimitrisPapail · DimitrisPapail · DimitrisPapail