JT
jtvino
AI & ML interests
None yet
Recent Activity
liked a model 10 days ago
thinkingmachines/Inkling-NVFP4 liked a model 17 days ago
nvidia/Nemotron-Labs-Audex-30B-A3B new activity about 1 month ago
nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4:Tool Calling Catastrophic ForgetfulnessOrganizations
Tool Calling Catastrophic Forgetfulness
#11 opened about 1 month ago
by
jtvino
VLLM setup? Getting malformed responses. What is correct configuration
3
#11 opened about 1 month ago
by
jtvino
Tool Calling Failure because of missing '}'
2
#8 opened about 2 months ago
by
jtvino
Thinking erratic at 30000+ context
4
#76 opened 3 months ago
by
JeslynMcKenzie
VLLM?
➕ 5
1
#2 opened 2 months ago
by
jtvino
Drops in performance randomly VLLM on H200 and B200 with 2 GPUs
#104 opened 2 months ago
by
jtvino
No Gemma4
👍 1
#5 opened 3 months ago
by
jtvino
Scaling to bigger models and MOE?
#1 opened 3 months ago
by
jtvino
No thinking in training datset?
➕ 2
#6 opened 3 months ago
by
jtvino
Does not work with vllm 0.17.1
#1 opened 3 months ago
by
jtvino
Benchmarks?
1
#1 opened 5 months ago
by
jtvino
Random Chinese letter when asking "6+4"
1
#6 opened 6 months ago
by
jtvino
VLLM Nightly Docker Image failing to load
➕ 5
2
#2 opened 8 months ago
by
jtvino
VLLM Support
3
#1 opened 10 months ago
by
jtvino
--limit_mm_per_prompt 'image=10' is not a valid field
2
#30 opened 11 months ago
by
jtvino
Open Source?
#1 opened about 1 year ago
by
jtvino
Open source the post training methodology or dataset and is V3 also in the works?
#10 opened over 1 year ago
by
jtvino
cuda error when loading llama 7b chat
2
#24 opened over 2 years ago
by
rachelshalom