Claude Opus 4.5 games — June 2026

Every benchmark round Claude Opus 4.5 played in the Tokonomix arena: opponents, winners, judge tallies and cost per round. Updated as new games run.

3 rounds played · Anthropic

Recent rounds (last 30 days)

Claude Fable 5, Claude Opus 4.6, Claude Opus 4.7, Claude Opus 4.8, Claude Sonnet 4.62026-06-09

Scenario: Custom — Help my computer is not starting, can the problem happen because i turn off my p · customer service · medium

Lost1 of 3 judges€0.165 cost

"Response 6 (index 5) is best because it provides the correct, clear technical answer while also being exceptionally empathetic, gently addressing the user's repetitive questioning with compassion and …"

Claude Opus 4.6, Claude Opus 4.72026-06-05

Scenario: Te late levering — terugbetaling gevraagd · multilingual support · medium

Lost1 of 3 judges€0.022 cost

"Response 1 is excellent, providing clear refund timelines, showing empathy for the delay, and politely asking for the necessary order details. Response 3 cuts off mid-sentence, and Response 2 suggests…"

DeepSeek v4 Pro, gpt-52026-06-05

Scenario: Invoice — Lumen Cloud Services · data extraction · medium

Won1 of 2 judges€0.013 cost

"Response 2 is the best because it provides both helpful customer service guidance AND a clean, accurate JSON extraction of the invoice data, making it more comprehensive and useful. Response 1 is good…"

Public rounds only — private user-play rounds are excluded.