Hacker News
new
|
past
|
comments
|
ask
|
show
|
jobs
|
submit
login
skirmish
on Oct 2, 2024
|
parent
|
context
|
favorite
| on:
AMD GPU Inference
They prefer you pay $3,600 for AMD Radeon Pro W7900, 48GB VRAM.
anthonix1
on Oct 3, 2024
[–]
... which also has a much lower power cap
skirmish
on Oct 3, 2024
|
parent
[–]
Not that much lower, 295W vs 355W, and for LLM inference VRAM bandwidth is the main bottleneck. But the price is ridiculous.
Guidelines
|
FAQ
|
Lists
|
API
|
Security
|
Legal
|
Apply to YC
|
Contact
Search: