Hacker News
new
|
past
|
comments
|
ask
|
show
|
jobs
|
submit
login
mirekrusin
27 days ago
|
parent
|
context
|
favorite
| on:
How Taalas “prints” LLM onto a chip?
gpt-oss is fp4 - they're saying they'll next try mid size one, I'm guessing gpt-oss-20b then large one, i'm guessing gpt-oss-120b as their hardware is fp4 friendly
Guidelines
|
FAQ
|
Lists
|
API
|
Security
|
Legal
|
Apply to YC
|
Contact
Search: