Park Soobin

subin

vllm-sr

·

https://cv.subin.rs

AI & ML interests

LLM inference and serving — vLLM internals, chat templates, tokenization, and NPU/GPU deployment.

Recent Activity

new activity 10 days ago

vllm-sr/d3-ultra-FP8:Is d3-ultra-FP8 too heavy for production concurrency?

liked a model 10 days ago

baidu/Unlimited-OCR-2

published a model 11 days ago

vllm-sr/d3-ultra-INT4

View all activity

Organizations

upvoted a collection 17 days ago

📡 Vela 3.0

Six open routing models, 0.3B to 27B: the router's decisions and the spans you name, in one call. • 10 items • Updated 1 day ago • 15

upvoted 3 collections about 1 month ago

Decision models

GGUF decision models for the /v1/systemone API in llama.cpp • 16 items • Updated about 1 month ago • 41

🔮Decision 3.0

Towards Open Multimodal Foundation Decision Models • 9 items • Updated about 5 hours ago • 140

🧙‍ Decision 2.0

Towards Open Foundation Decision Models • 8 items • Updated Oct 11 • 99

upvoted a collection about 2 months ago

⛵ Vela 1.0

The model family for intelligent routing. • 15 items • Updated Oct 1 • 28