N
peonist
AI & ML interests
None yet
Recent Activity
updated a model about 3 hours ago
peonist-ai/halogen-npu-decider-4b updated a model about 3 hours ago
peonist-ai/halogen-npu-decider-0.8b updated a model about 18 hours ago
peonist-ai/halogen-qwen3.8-flash-nextOrganizations
The Best REAP version of Qwen3.6-35B-A3B
👍 1
1
#3 opened 3 days ago
by
budkovsky
Broken tool calls since ~0.17.0
2
#13 opened 4 days ago
by
rekillkos
Can I create my own quantization of Qwen3.8 using `hgn` format?
8
#12 opened 13 days ago
by
auf1r2
Very fast.
36
#1 opened about 1 month ago
by
theTross
Thank you!
❤️ 1
#1 opened 13 days ago
by
peonist
Halogen Server 4-bit Inference Question
1
#11 opened 13 days ago
by
markmonger
Model keeps assuming key points are a typo (general knowledge)
2
#2 opened 15 days ago
by
Algiefonso
Ablated model?
41
#4 opened about 1 month ago
by
haign
Q5 or Q6 quants ?
7
#6 opened 29 days ago
by
auf1r2
Variable MTP and ngram
1
#9 opened 21 days ago
by
davenetdev
What an absolute godlike tool. The speed is insane!
❤️ 11
25
#7 opened 26 days ago
by
DarkCobalt
Can the model size be further reduced to lower initial loading overhead
4
#8 opened 24 days ago
by
yhx03
Donato's video on halogen & qwen3.8 flash
👍❤️ 9
2
#5 opened about 1 month ago
by
iamtc
How about a 3-bit quant?
8
#3 opened about 1 month ago
by
PhilTomson
Engram SSD Offloading
4
#2 opened about 1 month ago
by
justanonymuser
Very Good Quant
3
#1 opened about 1 month ago
by
DarkKitsune
Very fast.
36
#1 opened about 1 month ago
by
theTross
Very Good Quant
3
#1 opened about 1 month ago
by
DarkKitsune
MTP layers
3
#26 opened 3 months ago
by
Treene
MTP
🔥➕ 17
3
#2 opened 4 months ago
by
engrtipusultan