Kevin Gascoigne

Kjay

·

AI & ML interests

Graphic Design

Recent Activity

new activity about 13 hours ago

nvidia/Qwen3.6-27B-NVFP4:Works via Ollama, but the ~40 tok/s MTP speedup requires raw llama-server (Ollama has no --spec-type flag) - plus vLLM lockup findings on DGX Spark

new activity 22 days ago

joaquinpedreno/hello_world:Local LLM option?

liked a model 29 days ago

nhathoangfoto/Flux.2-Klein-9B-SmartCharacterSwap

View all activity

Organizations

New activity in nvidia/Qwen3.6-27B-NVFP4 about 13 hours ago

Works via Ollama, but the ~40 tok/s MTP speedup requires raw llama-server (Ollama has no --spec-type flag) - plus vLLM lockup findings on DGX Spark

#11 opened 1 day ago by

New activity in joaquinpedreno/hello_world 22 days ago

Local LLM option?

#1 opened 22 days ago by

New activity in Phr00t/Qwen-Image-Edit-Rapid-AIO 8 months ago

v9 Testing

#102 opened 8 months ago by

New activity in TIGER-Lab/ImagenWorld 9 months ago

Thank you.

#2 opened 9 months ago by

New activity in JD3GEN/JD3_Nudify_Kontext_LoRa 12 months ago

Non Consistent Results

#5 opened 12 months ago by

New activity in ShineChen1024/MagicClothing almost 2 years ago

I always got this error message, How to solve it? pls...

#2 opened about 2 years ago by