Ideas & society2023 · shared on Hacker News
The case for 4-bit precision: k-bit Inference Scaling Laws
Explore The case for 4-bit precision: k-bit Inference Scaling Laws, 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.
Yalbert_e ·
Meta AI releases Code Llama 70B
598 points294 comments
Ybrylie ·
Gemma3 – The current strongest model that fits on a single GPU
252 points138 comments
Yb_mc2 ·
Brave Leo now uses Mixtral 8x7B as default
244 points178 comments
Yrbanffy ·
AMD now has more compute on the top 500 than Nvidia
198 points134 comments
Ymsoad ·
Llama.cpp 30B runs with only 6GB of RAM now
124 points26 comments
Ystared ·
Qwen 3.6 27B is the sweet spot for local development
117 points68 comments
Ytosh ·
Qwen1.5-110B
114 points58 comments
Yog_kalu ·
Alpaca- An Instruct Tuned Llama 7B. Responses on par with txt-DaVinci-3. Demo up
53 points17 comments
Ycocktailpeanut ·
Dalai: Automatically install, run, and play with LLaMA on your computer
5 points1 comments
Ybirriel ·
How Is LLaMa.cpp Possible?
4 points0 comments
Ybackpropaganda ·
Some Remarks on Large Language Models – Yoav Goldberg
3 points0 comments
Yanigbrowl ·
QUIK is a method for quantizing LLM post-training weights to 4 bit precision
3 points0 comments
Ydebdut ·
Visual ChatGPT
2 points1 comments
Ymrtksn ·
LLaMA-7B in Pure C++ with full Apple Silicon support
2 points1 comments
YPalmik ·
The case for 4-bit precision: k-bit Inference Scaling Laws
1 points0 comments