Next.js Hacker News
top
|
new
|
ask
|
show
|
jobs
|
GitHub
SMLL: Using 200MB of Neural Network to Save 400 Bytes
14 points by
fcjr
13 hours ago |
2 comments
add comment
msephton
No mention of decompression speed and validation, or did I miss something?
savalione
It's in the post: Benchmarks -> Speed
tl;dr: SMLL is approximately 10,000x slower than Gzip