Y
Hacker News
new
|
ask
|
show
|
jobs
by
not_that_d
196 days ago
What is "Vibe testing"?
2 comments
catigula
196 days ago
He means capturing things that benchmarks don't. You can use Claude and GPT-5 back-to-back in a field that score nearly identically on. You will notice several differences. This is the "vibe".
link
BizarroLand
196 days ago
I would assume that it is testing how well and appropriately the LLM responds to prompts.
link