Hacker News .hn
new
|
past
|
comments
|
ask
|
show
|
jobs
|
submit
login
jgalt212
25 days ago
|
parent
|
context
|
favorite
| on:
GLM-5.2 is the new leading open weights model on A...
> as input tokens that are already in the KV cache are practically free for the provider,
not at today's RAM prices.
stymaar
25 days ago
[–]
RAM price don't change anything. You can't fit an infinity of tokens in the KV cache, but the ones that are in there when a user request them are still practically free.
jgalt212
24 days ago
|
parent
[–]
so the utility of the KV cache has nothing to do with available RAM?
stymaar
23 days ago
|
root
|
parent
[–]
This proposition has no logical link with any of the above. Read again.
Guidelines
|
FAQ
|
Lists
|
API
|
Security
|
Legal
|
Apply to YC
|
Contact
Search:
not at today's RAM prices.