Is Opus Really 3rd-4th Best? Here's What People Think

Not according to a recent Twitter poll. @striedinger's claim that Opus is only the 3rd or 4th best model sparked debate, and the results are pretty interesting.

Here's the breakdown:

Some users had strong opinions. For instance, @ayushporwalhq said, "What are your top 3? In my experience, nothing comes close to opus. Curious to try your top 3." Another user, @proxy_vector, noted, "Depends heavily on the task tho. For creative writing and nuanced reasoning Opus still has an edge that benchmarks dont capture. Rankings are kinda me..."

One bold claim came from @DanielWhit21874: "Yep because they’re simply better models out there for instance GPT 5.2 Codex high 👀"

While the sample size is small, and "Opus is not the 3rd-4th best" got the most votes, the poll suggests that many users believe Opus is a top-tier model, possibly even the best, particularly for specific tasks like coding. Although "Opus is the best" didn't get a ton of votes, its supporters expressed intense passion.

It seems like the perceived value of Opus really depends on individual use cases and expectations, showing that model rankings are always subjective.