Hacker News
new
|
past
|
comments
|
ask
|
show
|
jobs
|
submit
login
londons_explore
3 months ago
|
parent
|
context
|
favorite
| on:
Fooling around with encrypted reasoning blobs
If it actually cost that much RAM, they would almost certainly add extra things to the API to manage cache lifetime. Ie. A 'please cache this for X minutes' flag, or a setting for a single re-use cache (the most common use case)
cyanydeez
3 months ago
[–]
https://platform.claude.com/docs/en/build-with-claude/prompt...
suggests the can cache outside the gpu.
Guidelines
|
FAQ
|
Lists
|
API
|
Security
|
Legal
|
Apply to YC
|
Contact
Search:
