pottokao/Qwen3.8-27B-NVFP4-MTP-2x16GB
Text Generation β’ 19B β’ Updated β’ 304 β’ 2
Val
curiouspp8
Β·
AI & ML interests
None yet
Recent Activity
liked a model 1 day ago
pottokao/Qwen3.8-27B-NVFP4-MTP-2x16GBOrganizations
None yet
liked a
model 1 day ago
Vision encoder crashes with Encoder cache miss on TP2 (without --language-model-only)
3
#1 opened 2 days ago
by curiouspp8
TP2 benchmark on 2x RTX 5070 Ti (16GB) -- vision works, decode speed + VRAM findings
π 2
#16 opened 2 days ago
by curiouspp8
Testing IQ3_K
π₯ 1
8
#6 opened 4 months ago
by 
shewin
How to use mmproj
π 1
1
#8 opened 4 months ago
by curiouspp8
smol-IQ2_KL bench, GPUs
π₯ 1
5
#5 opened 4 months ago
by curiouspp8
GLM 5.1 vs GLM 5 - burns A LOT output tokens on thinking
π 1
35
#6 opened 4 months ago
by curiouspp8
Highest performance inference on <8 RTX 6000 Pros setups
2
#1 opened 4 months ago
by curiouspp8
Quick bench for smol-IQ3_KS on 2 GPUs
π 1
7
#4 opened 4 months ago
by curiouspp8
Quick bench for IQ2_KS on 1 GPU
π 1
5
#3 opened 4 months ago
by curiouspp8
What quant would be closest to AWQ?
1
#5 opened 4 months ago
by curiouspp8
4bpw request =)
π 1
13
#2 opened 4 months ago
by 
BahamutRU
Testing IQ3_KS
π 1
10
#3 opened 4 months ago
by shewin
Speed inference UD-IQ2_M
π€― 1
1
#2 opened 4 months ago
by 
Ukro
Highest performance inference on <8 RTX 6000 Pros setups
#6 opened 4 months ago
by curiouspp8
Typical GLM 5.1 overhead on top of weights memory
π 1
4
#2 opened 4 months ago
by curiouspp8
1x RTX 5090 on EPYC, CPU-only speeds
β€οΈ 3
11
#5 opened 6 months ago
by 
sousekd
1x RTX 5090 on EPYC, CPU-only speeds
β€οΈ 3
11
#5 opened 6 months ago
by sousekd