Hacker News
new
|
past
|
comments
|
ask
|
show
|
jobs
|
submit
login
zozbot234
15 days ago
|
parent
|
context
|
favorite
| on:
DeepClaude – Claude Code agent loop with DeepSeek ...
Waiting for official support in llama.cpp. There is a fork that can run a lightly quantized (Q2 expert layers) DeepSeek V4 Flash in 128GB RAM without offloading weight fetches from disk.
karel-3d
15 days ago
[–]
Ouch. Can't run that on my M4 mac with 48GB RAM.
Guidelines
|
FAQ
|
Lists
|
API
|
Security
|
Legal
|
Apply to YC
|
Contact
Search:
