Is the opus 5.5 hype legit?

A user on Reddit questioned the hype surrounding Opus 5.5, noting that while many claim it "can really code," models like Opus 4.5 and GPT 5.2 have been proficient at coding for some time. The user, who has switched between Claude Code, Codex, and GPT models, finds that newer versions simply require less oversight. They also mentioned being on a $200 plan for these services.

This user's direct experience with Opus 5.5, Claude Code, and GPT 5.4 offers a comparative perspective on coding capabilities, unlike general hype.