Qingyun Li

PyTorch-Geometric 2.8.0- base: Torch-2.8.0, PyG-2.8.0 wheels, Java

Jetson Orin Nano Super image combining mamba_ssm and opencv via dusty-nv/jetson-containers

Llama Factory, CUDA 12.8, uv

vLLM 0.10.1.1 with Python 3.11, CUDA 12.8, PyTorch 2.8, uv, huggingface_hub, datasets.