noonghunna/club-3090
Community recipes for serving LLMs on RTX 3090. Multi-engine (vLLM, llama.cpp, SGLang) and model-agnostic. Currently shipping Qwen3.6-27B configs for 1× and 2× cards.
Language: Shell
Stars: 457 Issues: 6 Forks: 24
https://github.com/noonghunna/club-3090
Post #13244
1.61K