The most prestigious law school admissions discussion board in the world.
|
 |
 
|
SWE-bench is the only measure that useful and it's still poor
| https://i.imgur.com/YbMiOBE.png | 09/13/26 |
Poast new message in this thread
Date: September 13th, 2026 8:26 PM Author: https://i.imgur.com/YbMiOBE.png These fags always test a mix of models running at different quants with different tunings. Try using your eyes and your brain to evaluate a model HTH (http://www.autoadmit.com/thread.php?thread_id=5903653&forum_id=2)#50134285) |