Help improve this record. Suggest a correction
AMD · indexed by the LabUpstream recipeA local model server on Ryzen AI Max
AMD’s gpt-oss-20b example connects a GGUF model to llama.cpp with ROCm acceleration.
THE SETUP
An upstream starting point
The vendor guide supplies a prebuilt runtime and a local server example with 2,048-token context, GPU offload and flash attention. It is a software starting point for compatible Ryzen systems.
Open AMD’s llama.cpp guideWhat supports this record
AMD documentation indexed 10 September 2026. The Lab has not reproduced it on Halo, Framework, HP or MINISFORUM hardware.
THE KNOWLEDGE AROUND IT