Alexone
krustik
AI & ML interests
Testing Evaluating
Recent Activity
new activity
6 days ago
zai-org/GLM-5:Great Model but not accessible anymore
new activity
9 days ago
stepfun-ai/Step-3.5-Flash:Question : Real-world use cases for Step-3.5-Flash
new activity
9 days ago
unsloth/Qwen3-Coder-Next-GGUF:stepfun 3.5 flash pls
Organizations
None yet
Great Model but not accessible anymore
โ
๐ฅ
4
8
#17 opened 7 days ago
by
darkstar3537
Question : Real-world use cases for Step-3.5-Flash
8
#24 opened 10 days ago
by
Geodd
stepfun 3.5 flash pls
2
#7 opened 13 days ago
by
semon017
error 403
1
#1 opened 12 days ago
by
krustik
Great progress but pretty unusable in the real world still
4
#24 opened 23 days ago
by
dynjo
hi, who are you? -> I am Claude.
6
#59 opened 15 days ago
by
Jeol
high hallucinations
1
#1 opened about 1 month ago
by
krustik
Request Details on GPU ad memory requirements
5
#16 opened about 2 months ago
by
DragoZatch
llama cpp
๐ฅ
4
3
#8 opened about 2 months ago
by
moon005
GLM-4.7-Q8_0.gguf capabilities with Ollama
13
#5 opened about 2 months ago
by
myfenris
quantization damaged coding
4
#2 opened about 2 months ago
by
krustik
Is the GLM series a coding only model now?
๐
1
7
#2 opened about 2 months ago
by
Doctor-Chad-PhD
BF16 or Q8_K_XL - which would give more accurate coding results?
5
#6 opened 2 months ago
by
TimothyRoo
Fits in RTX 3090, which is best time to buy ($500 on used market)
#18 opened 2 months ago
by
krustik
Super broken, but i didnt expect much
๐
1
1
#1 opened 2 months ago
by
rombodawg
Quantization destroyed coding abilities in Q5 (735Gb RAM)
17
#4 opened 3 months ago
by
krustik
Beware quantization destroyed coding abilities, broken code in Q5-K-M (735Gb RAM)
10
#21 opened 3 months ago
by
krustik
mixed results
2
#28 opened 4 months ago
by
kingriel
Can it run on NVIDIA A100 80G * 8 ๏ผ
2
#30 opened 4 months ago
by
jetto98
Failed test of restoring QR code image vs flux Kontext success
1
#16 opened 4 months ago
by
krustik