/ollama_tegra_fix

Quick ollama hack to troubleshoot Tegra issues

Primary LanguageDockerfile

Ollama Tegra Fix

Jetson support should be included as of Ollama 0.1.30. Please see the official repo for more information https://github.com/ollama/ollama

ENSURE THESE TWO VARIABLES ARE SET:

export LD_LIBRARY_PATH="/usr/local/cuda/lib64:/usr/local/cuda/compat:/usr/local/cuda/include" export OLLAMA_SKIP_CPU_GENERATE="1"

Also, Tegra devices will fail go generate ./... unless architectures are set:

L4T_VERSION.major >= 36: # JetPack 6

export CMAKE_CUDA_ARCHITECTURES="87"

L4T_VERSION.major >= 34: # JetPack 5

export CMAKE_CUDA_ARCHITECTURES="72;87"

L4T_VERSION.major == 32: # JetPack 4

export CMAKE_CUDA_ARCHITECTURES="53;62;72"