New Weights working on vllm DGX Spark -1 node setup - GOOD QUALITY
17
#25 opened 11 days ago
by 
TeAmErX
Bug report: Long form generation collapses into word-salad recitation
#24 opened 12 days ago
by 
squawksquack
New Weights Break vllm DGX Spark setup
4
#23 opened 14 days ago
by 
park3y
the new update break vllm with one RTX Pro 6000 Blackwell
👍 6
4
#22 opened 14 days ago
by 
tangram7
2x DGX Spark with main branch
9
#21 opened 15 days ago
by 
karypid
DGX Spark
👀 1
10
#20 opened 16 days ago
by 
coe-arrtificial
My Experience so far (great for Hermes Agent, dissappointing for Text to SQL )
#19 opened 18 days ago
by 
AlioLeuchtmann
Is there a recommended vLLM recipe for RTX Pro 6000?
3
#18 opened 20 days ago
by 
DikiyLifter92
RC2 is now too large for RTX 6000
😔 6
12
#17 opened 21 days ago
by 
ndurkee
Reproducible Laguna S 2.1 NVFP4 runaway reasoning: output budget scales 2K → 16K with 12/12 length exhaustion and no final answer
👀👍 8
7
#16 opened 22 days ago
by 
darkmatter2222
Reproducible Laguna S 2.1 / vLLM reasoning-parser mismatch causes malformed multi-turn history
👀 2
7
#15 opened 23 days ago
by darkmatter2222
Not thinking even when `chat_template_kwargs` is set?
👍 3
10
#13 opened 25 days ago
by 
evilperson068
My yaml if anyone wants it (DGX Spark + Docker + Portainer.io + Prometheus + Grafana)
🔥 1
#12 opened 25 days ago
by darkmatter2222
Still looping after update - details inside
👀 2
22
#10 opened 25 days ago
by 
cbert33
Reasoning budget?
4
#9 opened 26 days ago
by cbert33
Writes code in thoughts
#8 opened 26 days ago
by 
darkstar3537
looping..
👍 2
13
#7 opened 26 days ago
by 
remaxx
Looping :(
😔 3
6
#6 opened 26 days ago
by 
idleDaemon
Looping behavior
😔 1
1
#5 opened 26 days ago
by 
vladimir94
Pretty good :) but ... looping :(
5
#4 opened 27 days ago
by 
jbourny
Laguna-S-2.1-NVFP4-GGUF model
4
#1 opened 27 days ago
by 
lirex