Arena Compares Nine Claude Opus Outputs Across Model Versions
AI summary
Arena shared a video intended to show how Anthropic's Claude models have progressed over time. Peter Gostev generated nine outputs using the same prompt, spanning Claude Opus 4.5 through Claude Opus 5.5 (Max), and invited users to try more prompts on Arena.
In-depth insight
The same-prompt comparison offers users a direct visual way to examine output differences across Claude Opus versions, although the post provides little methodological detail.
Watch how @AnthropicAI’s models have progressed over time.
@petergostev generated nine outputs from Opus 4.5 to Opus 5.5 (Max) using the same prompt.
For more prompts to try on Arena, see the link below. https://x.com/arena/status/2123436909120856466/video/1