**Title:** NIM qwen3.5-35b-a3b:1.7.0-variant fails on Jetson AGX Thor — Triton ptxas-blackwell does not recognize sm_110a

[AastaLLL]

Hi,

Just FYI.
There is a sharing for NVIDIA-Nemotron-3-Super-120B-A12B-NVFP4 model

Running NVIDIA-Nemotron-3-Super-120B-A12B-NVFP4 on the Nvidia Jetson Thor

I just managed to get NVIDIA-Nemotron-3-Super-120B-A12B-NVFP4 running locally using my vLLM inference engine on a single NVIDIA Jetson AGX Thor with 128 GB of unified memory. 1. Create a fresh venv uv venv .vllm --python 3.12 source .vllm/bin/activate 2. Install torch with CUDA 13 uv pip install torch torchvision torchaudio \ --index-url https://download.pytorch.org/whl/cu130 3. Set environment for sm110 export TORCH_CUDA_ARCH_LIST=11.0a export CUDA_HOME=/usr/local/cuda-13 export TRITON_PTXA…

It build vLLM from the source directly.

Thanks.