ko
mirek190
AI & ML interests
None yet
Recent Activity
liked
a model
13 days ago
deepseek-ai/DeepSeek-R1-0528
new activity
about 1 month ago
bartowski/Qwen_Qwen3-30B-A3B-GGUF:Worse than expected performance
Organizations
None yet
mirek190's activity
Worse than expected performance
π
1
7
#1 opened about 1 month ago
by
martijk1
Perplexity - google_gemma-3-27b-it-qat-Q4_K_M has the worst score of all Q4 quants. The rest test in the thread.
#3 opened about 2 months ago
by
mirek190
R1 32b is much worse than QwQ ...
22
#6 opened 5 months ago
by
mirek190
How to trigger thinking?
5
#1 opened 6 months ago
by
mirek190
Misleading advertisement as reasoning model
π
5
8
#5 opened 7 months ago
by
esozbek

Can you provide code for inference with MCTS?
π
6
6
#3 opened 7 months ago
by
sszymczyk
9.11>9.9π
π
3
1
#6 opened 7 months ago
by
cydxg
Possibly the provided prompt format is wrong.
π
π§
6
12
#1 opened 9 months ago
by
vevi33
prompt template is wrong
10
#2 opened 10 months ago
by
mirek190
gguf quant please!!
π
π
10
7
#3 opened 10 months ago
by
gopi87
surprising - that model is incredible for it's size.
π
2
20
#2 opened 10 months ago
by
mirek190
Performance
π
1
36
#1 opened 11 months ago
by
urtuuuu
The first GGUF that works with long context on llama.cpp!
3
#1 opened 11 months ago
by
ubergarm
And where is the GGUF file itself?
β€οΈ
2
12
#1 opened 11 months ago
by
Anonimus12345678902
Gemma-2-9B-it scores
π
3
8
#843 opened 11 months ago
by
saishf

GGGGGGGGGGGGGGGGGGGGGGGGGGGG
9
#2 opened 11 months ago
by
x4k
whats this model do ? uncensored version ?
π
1
5
#1 opened 11 months ago
by
gopi87
For size of 9b and coding that model is IMPRESSIVE. How can be so good?
11
#1 opened 11 months ago
by
mirek190
For size of 9b and coding that model is IMPRESSIVE. How can be so good?
11
#1 opened 11 months ago
by
mirek190