Tutorial: Build llama.cpp from source and run Qwen3 235B

[_cjg]

@RazielAU

You shouldn’t have to set that parameter at all

Yes, that’s what I thought. At least that the CUDA environment is cleanly integrated. But not in the recovery image (dgx-spark-recovery-image-1.105.17.tar.gz) and all updates, where it seems to have been forgotten. One gets the feeling that the NVIDIA DGX department is a little too vain for Spark?
So, for our weight watchers spark self-help group (who has the same problem with the recovery image), you could add:

@aniculescu

Without knowing more to attempt a repro I’m not certain what the problem is, however, you could try putting
CUDACXX=/usr/local/cuda-13.0/bin/nvcc
into /etc/environment so that it is applied when using sudo.