Hacker News
new
|
past
|
comments
|
ask
|
show
|
jobs
|
submit
login
imiric
22 days ago
|
parent
|
context
|
favorite
| on:
Claude Opus 4.5
Or, 2b: the nerf is real, but benchmarks are gamed and models are trained to excel at them, yet fall flat in real world situations.
metalliqaz
22 days ago
[–]
I mostly stay out of the LLM space but I thought it was an open secret already that the benchmarks are absolutely gamed.
Guidelines
|
FAQ
|
Lists
|
API
|
Security
|
Legal
|
Apply to YC
|
Contact
Search: