John Leimgruber III
ubergarm
AI & ML interests
Open LLMs and Astrophotography image processing.
Recent Activity
liked
a model
about 2 hours ago
rednote-hilab/dots.llm1.inst
new activity
about 2 hours ago
ubergarm/DeepSeek-R1-0528-GGUF:1.5 bpw
new activity
about 2 hours ago
anikifoss/DeepSeek-R1-0528-DQ4_K_R4:Wahoo thanks for sharing your work!
Organizations
None yet
ubergarm's activity
1.5 bpw
โ
1
12
#6 opened 6 days ago
by
lmganon123
Wahoo thanks for sharing your work!
19
#1 opened 5 days ago
by
ubergarm
benchmarks
๐
1
2
#8 opened 1 day ago
by
BernardH
Thanks for your work! Any chance for something between Q2_K_R and Q3_K_R?
๐
๐
2
2
#7 opened 1 day ago
by
Panchovix
Interesting comparisons!
11
#1 opened 21 days ago
by
ubergarm
https://huggingface.co/blog/autoround
1
#1 opened 9 days ago
by
21world

Local Installation Video and Testing - Step by Step
โค๏ธ
2
4
#5 opened 6 days ago
by
fahdmirzac

Larger quants request
3
#4 opened 6 days ago
by
sousekd

Recommendation for 256 ram 48 vram
โค๏ธ
2
5
#2 opened 7 days ago
by
ciprianv
Fix file sizes
1
#3 opened 6 days ago
by
fuckoffIjustwantdown
benchmarks
๐
1
2
#3 opened 11 days ago
by
BernardH
Oops nevermind
๐
1
1
#1 opened 18 days ago
by
ubergarm
Q6 and Q8 quants please?
1
#1 opened 20 days ago
by
dipper9876
Any chance for IQ3_XXS/IQ3_XS or similar size?
3
#2 opened 20 days ago
by
Panchovix
Q6 K
36
#1 opened about 1 month ago
by
Autumnlight
Are there any changes in method vs. your v3-0324 quant?
3
#1 opened 23 days ago
by
usrlocalben
Ik_llama.cpp speed improvements over llama.cpp
๐ฅ
1
3
#3 opened 27 days ago
by
ciprianv
Simple --override-tensor
โค๏ธ
1
1
#4 opened 23 days ago
by
Kirara702
Please consider creating ik_llama.cpp compatible quants (without llama.cpp-specific MLA tensors)
1
#1 opened about 1 month ago
by
Lissanro

Context Length
3
#2 opened about 1 month ago
by
vacekj