JM
akierum
·
AI & ML interests
None yet
Organizations
None yet
What speed do you get at Q8 on AMD Ryzen™ AI Max+ 395
11
#14 opened 5 months ago
by
akierum
Where is vision support like previous devstral version?
#1 opened 3 months ago
by
akierum
New Chat Template + Tool Calling Fixes as of 05 Aug, 2025
👀 ❤️ 23
48
#10 opened 7 months ago
by
shimmyshimmer
Will there be fix new proper model or jinja template fix for Roocode/Cline/Goose etc.?
#14 opened 4 months ago
by
akierum
Well I tried the new DevQuasar Kwaipilot.KAT-Dev.Q8_0.gguf - Lmstudio and Jan.ai errors out:
6
#9 opened 5 months ago
by
akierum
This one's great!
6
#1 opened 5 months ago
by
kbuettner
the instruct following is poor
3
#9 opened 4 months ago
by
pypry
How much VRAM is required for this one
1
#1 opened 4 months ago
by
deleted
KAT-DEV 33B no longer working with Jan.ai 0.7.2 - Cline / LMstudio v0.3.31b4 - Cline
#12 opened 4 months ago
by
akierum
Is KAT-Dev supposed to be that much slower than QWEN?
4
#11 opened 4 months ago
by
deleted
KAT-DEV Q4 - Jan.ai - Roocode / Cline does not work Lmstudio crashes too (SOLVED)
1
#1 opened 4 months ago
by
akierum
coder vs yoyo-v4 vs thinking vs instruct in compex programming task
3
#2 opened 5 months ago
by
NIK2703
Request GGUF: Kwaipilot/KAT-Dev (62.4% (!) on SWE-Bench Verified having just 32B)
4
#15 opened 4 months ago
by
Reverger
There is new KAT-Dev-72B-Exp
9
#1 opened 5 months ago
by
akierum
This model gives errors in cline, but new official one does not
👀 1
3
#3 opened 5 months ago
by
akierum
why Kat-Dev worse then Qwen3-coder 30B?
12
#8 opened 5 months ago
by
akierum
errors out with roocode
3
#1 opened 5 months ago
by
akierum
running parameters?
👀 1
2
#5 opened 5 months ago
by
akierum
Tool Calling (jinja template) Issues
14
#1 opened 5 months ago
by
Bellesteck