Hacker News
new
|
past
|
comments
|
ask
|
show
|
jobs
|
submit
login
nkjvhb
23 days ago
|
parent
|
context
|
favorite
| on:
GLM-5.3-Flash
Lead doesn't really matter anymore. I just ported a very old cuda library to rocm, so it can be run on MI300s. 2 years ago this would have been a nightmare. Today it was an afternoon.
throwdbaaway
23 days ago
[–]
Exactly. Coding for inference is solved. CUDA is no longer a moat.
Guidelines
|
FAQ
|
Lists
|
API
|
Security
|
Legal
|
Apply to YC
|
Contact
Search: