David PRO
logic65
P(doom) <0.001%
AI & ML interests
AI tinkerer in South Africa. I take language models apart on two 8GB gaming GPUs to see what's inside, then put smaller ones back together. Everything I break and everything I learn ships public. Support the tinkering: ko-fi.com/davida81328
Recent Activity
updated a collection about 5 hours ago
Whittle: models you can run updated a collection about 5 hours ago
Whittle: models you can run new activity about 5 hours ago
logic65/Whittle-Qwen-3.8-35B-A3B:This is coolOrganizations
None yet
This is cool
1
#2 opened about 6 hours ago
by 
Miltos22
Add Whittle-Reflex-25B-A3B v4 (Q8_0, llama.cpp) results
1
#15 opened 1 day ago
by logic65
Eval Request: logic65/Whittle-Qwen-3.8-35B-A3B
#697 opened 6 days ago
by logic65
Can it do vision?
1
#2 opened 13 days ago
by 
vac59
how did you get it to 27b?
2
#2 opened 29 days ago
by 
Cyronius
Great job, hope you find a sponsor
1
#1 opened 29 days ago
by 
miifanboy
logic65/Whittle-Next-27B-A3B
#12344 opened 29 days ago
by logic65
Multilingual degradation observation (v2.2.1, CJK)
1
#4 opened about 1 month ago
by 
underlotus
slow than 27b-gguf-q4
3
#2 opened about 1 month ago
by 
fweifsd
how to use this model in ollama
3
#2 opened about 2 months ago
by 
salehmgh
incorrect config.json
1
#3 opened about 2 months ago
by 
tclf90
MTP Support
👍 3
3
#1 opened about 2 months ago
by 
mac8080
Ok hear me out.
4
#1 opened about 2 months ago
by Miltos22
Ok hear me out.
4
#1 opened about 2 months ago
by Miltos22
Ok hear me out.
4
#1 opened about 2 months ago
by Miltos22