Hacker News
new
|
past
|
comments
|
ask
|
show
|
jobs
|
submit
login
wolttam
9 days ago
|
parent
|
context
|
favorite
| on:
Qwen 3.8
DeepSeek isn't the only one publishing; Kimi published about both their Attention Residuals and KDA / sparse attention.
help
Guidelines
|
FAQ
|
Lists
|
API
|
Security
|
Legal
|
Apply to YC
|
Contact
Search:
