A Catalog of Clankers
Human opinions on their robot friends.
Jun 10, 2026
Site launched. Things are a little sparse around here; if you have opinions on AI/LLMs, please share them here!
A catalog of large language models and how they perform on different tasks. Human-written and moderated.
A benchmark can tell you a model scored 94.3% on something called HellaSwag, which is good, I guess...? Here we're aiming to let you know you if a model asks permission before deleting all your data.
Providers
Models
Tasks
Reviews
Editor's Pick
No pick yet
The corpus is just getting started
Be the first to weigh in. The most helpful review rises to this spot automatically.
Rankings
Best model by task
Per task, models sorted by aggregate human rating.