Back to Projects
notactuallytreyanastasio.github.io August 18, 2026 at 02:29 PM
Show HN: Shoehorn – Quantize any model down to run on your machine
40
Points
7
Comments
rhgraysonii
Author
Top Comments
hmokiguessAug 18
Reminds me of https://github.com/AlexsJones/llmfit
mbuchel-hnAug 18
does this work similar to airllm? i am wondering how it would handle something like quantizing kimi k3 on a budget of 8 gbs, or is that something you are not attempting to solve yet?
jaylaneAug 18
tried it out but based on the model sizing result i got i got an insufficient memory error when the server started running
Visit the Original Link
Read the full content on notactuallytreyanastasio.github.io
Source
notactuallytreyanastasio.github.io
Author
rhgraysoniiPosted
August 18, 2026 at 02:29 PM
More Top Stories
blog.plover.com Aug 14
Seven books I keep close because I love them
308135 commentsby surprisetalk
Details rustdesk.com Aug 14
RustDesk now supports true unattended remote access on Wayland
293 commentsby rustdesk
Details blog.google Aug 14
Google is making private AI practical with homomorphic encryption
497288 commentsby u1hcw9nx
Details