I put Fable 5.1 vs Opus 5.5 to the test on redesigning our in progress focus group session with live…

I put Fable 5.1 vs Opus 5.5 to the test on redesigning our in progress focus group session with live runtime insights. For a task requiring creativity and taste, I expected Fable to do better, but instead I was blown away with what Opus 5.5 suggested. Both models got the exact same prompt. Opus 5.5 created multiple versions in the canvas inside claude code (using the /design skill) and each widget it proposed was significantly better made when compared to Fable. I then asked Opus to create a video showing our brainstorming process - assembling the page and trying out different widgets in different places. It literally one shot the video (not perfect, but 80% there). Blow away with @AnthropicAI on this release. We're so back.