Help improve this record. Suggest a correction
NVIDIA · indexed by the LabUpstream recipeA GGUF model server on GB10
A source-based starting point for running llama.cpp on Spark and documenting your own configuration.
THE SETUP
An upstream starting point
NVIDIA provides a Spark-specific guide to building llama.cpp with GPU support and serving a model through an API. Use the original instructions for installation and supported settings.
Open NVIDIA’s llama.cpp playbookWhat supports this record
Upstream guide reviewed 9 September 2026. No UT run or performance measurement is claimed.
THE KNOWLEDGE AROUND IT