Hacker News
new
|
past
|
comments
|
ask
|
show
|
jobs
|
submit
|
from
login
1.
Jeff-Code: a 0.8B model makes Qwen 3.8-27B coding 47% faster, same pass rate
(
github.com/firelex
)
2 points
by
firelex
40 days ago
|
past
2.
Jeff – Jev-compatible 0.8B decision models, trained at home, ~30 ms
(
github.com/firelex
)
576 points
by
firelex
47 days ago
|
past
|
231 comments
Guidelines
|
FAQ
|
Lists
|
API
|
Security
|
Legal
|
Apply to YC
|
Contact
Search:
