Sok Ann Yap
sokann
AI & ML interests
None yet
Recent Activity
new activity about 10 hours ago
sokann/GLM-5.2-GGUF-2.244bpw:Pretty solid results with ik_llama and OpenCode! updated a model about 11 hours ago
sokann/GLM-5.2-GGUF-2.244bpw new activity about 20 hours ago
sokann/GLM-5.2-GGUF-2.244bpw:PPLX / BenchmarksOrganizations
None yet
Pretty solid results with ik_llama and OpenCode!
🔥 1
5
#2 opened 3 days ago
by
phakio
PPLX / Benchmarks
3
#1 opened 4 days ago
by
Dsturb
Qwen3.6 version.
3
#2 opened about 2 months ago
by
dotcmsmy
Holy shmoly this thing likes to overthink
1
#3 opened about 2 months ago
by
JohnPaulTheThird
Same again for Qwen3.6-27B
2
#2 opened 2 months ago
by
resynth
Great job!
🤗 1
4
#2 opened 2 months ago
by
ubergarm
Nice work thanks for more ik_llama.cpp quants!
4
#1 opened 3 months ago
by
ubergarm
Insight into the "weird" data.
145
#3 opened 4 months ago
by
espen96
why llama.cpp crashed when user first input prompt
23
#7 opened 4 months ago
by
rosspanda0
Qwen3.5 Tool Calling Template
6
#5 opened 4 months ago
by
anikifoss
smol-IQ2_XS is a damn good quant!
👍🔥 7
7
#2 opened 4 months ago
by
tarruda
Decent PPL with 100% IQ4_KSS
🔥 1
12
#3 opened 7 months ago
by
sokann
Any plans on creating ik quants for REAP models, please?
6
#9 opened 7 months ago
by
xakepp
GGUF quant
➕ 3
2
#3 opened 8 months ago
by
sm54
Prompts with /nothink in the calibration data
2
#2 opened 8 months ago
by
sokann
Performance on Gaming Rig
❤️🔥 1
9
#5 opened 9 months ago
by
ubergarm
GLM 4.6 imatrix
4
#1446 opened 8 months ago
by
sokann
imatrix.dat file
1
#2 opened 9 months ago
by
sokann
"found point XXX not on grid" with llama-quant on IQ1_S/IQ2_XXS/etc when using imatrix
1
#1 opened 10 months ago
by
sokann
Bits for embedding / lm-head / non expert layers
1
#1 opened 11 months ago
by
sokann