Opus 5.5 vs Sonnet 5.5 on the same video prompt

The same prompt run on Opus 5.5 (6 minutes, ~151k tokens) and Sonnet 5.5 (11 minutes, ~172k tokens), with the author preferring Sonnet's output.

agim𓍯

@aGim_asfArchived 6 days agoOriginally posted 2026.10.23

agim𓍯 @aGim_asfOriginally posted 2026.10.23

Open original on X

ran the same prompt on opus 5.5 and sonnet 5.5 sonnet looks really good here •Opus: 6 mins, ~151k tokens •Sonnet: 11 mins, ~172k tokens which one would you pick and is this the kind of output you'd expect from a model routing to fable 5.5?

This post quotes another post: x.com/agim_asf/status/2113588020322693595

x.com/agim_asf/status/2113594508218675300Archived 2026.10.24 10:01

Open original on X