Hacker News
new
|
past
|
comments
|
ask
|
show
|
jobs
|
submit
login
coder543
on Jan 9, 2024
|
parent
|
context
|
favorite
| on:
Mixtral 8x7B: A sparse Mixture of Experts language...
How many tokens/s? Which quantization? If you could test Q4KM and Q3KM, it would be interesting to hear how the M2 Max does!
teilo
on Jan 9, 2024
[–]
No quantization (8_0). The full 48GB model. As for token count, I haven't tested it on more than 200 or so.
pilotneko
on Jan 10, 2024
|
parent
[–]
Isn’t 8_0 8-bit quantization?
teilo
on Jan 11, 2024
|
root
|
parent
[–]
Sorry. That was a major brain fart. Yes. 8-bit quantization, and using 49G of RAM.
Guidelines
|
FAQ
|
Lists
|
API
|
Security
|
Legal
|
Apply to YC
|
Contact
Search:
