Hacker News
new
|
past
|
comments
|
ask
|
show
|
jobs
|
submit
login
causal
34 days ago
|
parent
|
context
|
favorite
| on:
Grok 4.6
Yeah as models get better, valid benchmarks become more "trust me bro".
Guidelines
|
FAQ
|
Lists
|
API
|
Security
|
Legal
|
Apply to YC
|
Contact
Search: