Hacker News
new
|
past
|
comments
|
ask
|
show
|
jobs
|
submit
login
Stevvo
11 days ago
|
parent
|
context
|
favorite
| on:
GPT-5.2
The variance is way too high for this test to have any value at all. I ran it 10 times, and each pelican on a bicycle was a better rendition than that, about half of them you could say were perfect.
golly_ned
11 days ago
|
next
[–]
Compared to the other benchmarks which are much more gameable, I trust PelicanBikeEval way more.
reply
getnormality
11 days ago
|
prev
[–]
Well, the variance is itself interesting.
reply
Guidelines
|
FAQ
|
Lists
|
API
|
Security
|
Legal
|
Apply to YC
|
Contact
Search: