Development

Peer Arena pits five models in a debate only one survives

development2025-12-28

Claim

The Peer Arena benchmark pits five LLMs in a survivor-style debate, with Opus 4.5 winning the most peer votes without ever voting for itself.

Reporting date: 2025-12-28 (not necessarily the event date)

Sources

Attribution and reporting linked by this record; links do not imply independent verification.

Welcome to December 28, 2025 (theinnermostloop.substack.com)

Knowledge graphopen in browser →

About

Peer Arena

Reported in

Welcome to December 28, 2025

Actor

Anthropic

Evidence for

Referenced by

Welcome to December 28, 2025

Cite or correct this record

Citation formatCitation text

Peer Arena pits five models in a debate only one survives
Innermost Loop KB — secondary record
Reporting date: 2025-12-28
KB permalink: https://nicholsn.github.io/innermost-loop-kb/developments/2025-12-28-peer-arena-opus-victor
Claim: The Peer Arena benchmark pits five LLMs in a survivor-style debate, with Opus 4.5 winning the most peer votes without ever voting for itself.
Attributed sources:
- Welcome to December 28, 2025 — https://theinnermostloop.substack.com/p/welcome-to-december-28-2025
This citation does not imply independent verification of linked-source claims.

Suggest a correction — opens a GitHub draft for you to review and submit.