I took a look and there are a wide array of different precisions in the "1-bit" quants. @danielhanchen , how can a "1-bit" model contain layers at Q8_0 and F32? Does 1-bit only refer to the quantization of the ffn* layers? In that case, what is the average precision of the entire "1-bit" model?
Michael
electrolanche
AI & ML interests
None yet
Recent Activity
new activity 5 days ago
sensenova/SenseNova-Loop-9B-Preview:What is RaccoonPPTBench1 ??? new activity 7 days ago
orcarouter/MiniMax-M3.1-Flash-abliterated-GGUF:256GB DRAM version? new activity 3 months ago
poolside/Laguna-S-2.1:BenchMAXXXEDOrganizations
None yet
What is RaccoonPPTBench1 ???
2
#3 opened 9 days ago
by 
NezTheNaughty
256GB DRAM version?
#1 opened 7 days ago
by electrolanche
BenchMAXXXED
➕ 2
33
#28 opened 4 months ago
by 
TawusGames
replied to danielhanchen's post 5 months ago
replied to danielhanchen's post 5 months ago
Sample size n=1...
But seriously, it's impressive that a binary tree can output code that even executes. And at 1-bit there is no dynamic quantization (by definition), can't go lower than 1-bit!
What is the difference between this model and the original google/gemma-4-E4B-it?
👀 1
3
#1 opened 6 months ago
by 
adel-cybral
replied to unmodeled-tyler's post about 1 year ago
I’m curious to see how it compares against it’s base model, Qwen3 8B.
updated 3
collections almost 2 years ago