Hacker News
new
|
past
|
comments
|
ask
|
show
|
jobs
|
submit
login
valine
on April 10, 2023
|
parent
|
context
|
favorite
| on:
The LLama Effect: Leak Sparked a Series of Open So...
Are you using quantization? I’ve gotten very good results from the float16 13B vicuna model.
tarruda
on April 10, 2023
[–]
Did you use GPU inference? If so, how much memory is required?
valine
on April 10, 2023
|
parent
[–]
Yeah I'm using GPU inference. The vicuna 13B model uses 26.3GB of VRAM with my setup. I'm running it split between two rtx4090s which gives me about 20 token/s.
Guidelines
|
FAQ
|
Lists
|
API
|
Security
|
Legal
|
Apply to YC
|
Contact
Search:
