Hacker Times
new
|
past
|
comments
|
ask
|
show
|
jobs
|
submit
login
ACCount37
26 days ago
|
parent
|
context
|
favorite
| on:
Claude Sonnet 5
Most benchmarks notoriously undercount multi-turn instruction following too. And it's where open source models (and Gemini) lose big time.
Guidelines
|
FAQ
|
Lists
|
API
|
Security
|
Legal
|
Apply to YC
|
Contact
Search:
