Ideas & society2023 · shared on Hacker News
QMoE: Practical Sub-1-Bit Compression of Trillion-Parameter Models
Explore QMoE: Practical Sub-1-Bit Compression of Trillion-Parameter Models, its original source and Hacker News discussions. Discover community perspectives and save it to your reading list.
GO BEYOND THE ABSTRACT
The conversation around the idea.
Original Hacker News threads. Different perspectives, questions, and things worth thinking about.
Showing up to 100 related threads. A paper may appear in a comment rather than the original submission.
Links found in the discussion
Up to 50 recent source links, in HN item order.
Yhomarp ·
Towards 1-bit Machine Learning Models
351 points157 comments
Yjnord ·
“Imprecise” language models are smaller, speedier, and nearly as accurate
302 points169 comments
Yjasondavies ·
QuIP#: 2-bit Quantization for LLMs
3 points0 comments
Ytitaniumtown ·
QMoE: Practical Sub-1-Bit Compression of Trillion-Parameter Models
3 points0 comments
YJCMais ·
QMoE: Practical Sub-1-Bit Compression of Trillion-Parameter Models
1 points1 comments