Re Re
bfkxnr
Β·
AI & ML interests
None yet
Recent Activity
new activity 8 days ago
bottlecapai/ThinkingCap-Qwen3.8-27B:Thoughts on SSM conv1d patch? new activity 8 days ago
bottlecapai/ThinkingCap-Qwen3.8-27B:use with DFLASH2 drafter possible? new activity 8 days ago
bottlecapai/ThinkingCap-Qwen3.8-27B-GGUF:Embeddings quantizationOrganizations
None yet
Thoughts on SSM conv1d patch?
π 4
#6 opened 8 days ago
by
bfkxnr
use with DFLASH2 drafter possible?
π 1
7
#2 opened 12 days ago
by
ghit72
Embeddings quantization
2
#6 opened 8 days ago
by
bfkxnr
GSQ-RCO quantization
π 8
3
#2 opened 12 days ago
by
Exo87
improve unsloth
3
#129 opened 23 days ago
by
l0rdraiden
New UD-Q8-K-L bigger than Q8_0 but less accurate?
2
#92 opened about 2 months ago
by
rosspanda0
Optimal settings for n-min and p-min?
π 2
3
#2 opened about 2 months ago
by
bfkxnr
Frankenstein in Five Pieces: How to Build a Music Model from Spare Parts, Duct Tape, and Desperate Prayers
β€οΈπ 3
38
#22 opened about 2 months ago
by
AdrienneNoctis
Froggeric just release the new Qwen-Fixed-Chat-Templates v22
β€οΈ 2
3
#2 opened about 2 months ago
by
lawlietr
Le Cirque du Raisonnement: 27B Parameters, 262K Tokens of Hot Air, and a Packaging Bug That Silently Truncates Every Prompt)
π₯β 5
19
#144 opened about 2 months ago
by
AdrienneNoctis
Reverse this string: '.DefaultCellStyle'
π₯ 3
21
#50 opened about 2 months ago
by
owao
can you make gguf?
6
#1 opened about 2 months ago
by
Gavin-chen
Code acceptance rate and token speed drops on write tool calls in pi coding agent
πβ 3
6
#3 opened about 2 months ago
by
sandeshrajx
Why Qwen3.8-27B overthinks? Here the reason and partial fix confirmed by benchmarks.
β€οΈπ 25
29
#76 opened about 2 months ago
by
LuffyTheFox
Horrible model .... Overthink eats all context tokens
βπ 6
23
#136 opened about 2 months ago
by
kremerneil
Is this actually UD quantization?
π 1
#1 opened about 2 months ago
by
bfkxnr
Definitely helps!
1
#2 opened about 2 months ago
by
bfkxnr