Skip to content

armgit / ODS

Public Imported from GitHub
0

ODS V3 Pre-Release: Public testing and refinement ahead of the official V3 launch. Turn your PC, Mac, or Linux box into a private AI server.

411 B
Raw Blame
# /etc/modprobe.d/amdgpu_llm_optimized.conf — GTT memory for LLM inference
# Install: sudo cp this /etc/modprobe.d/amdgpu_llm_optimized.conf && sudo update-initramfs -u
# Requires BIOS UMA Frame Buffer set to minimum (512 MB or 1 GB)
# HIP allocates from GTT on Strix Halo — this is where model weights live
options amdgpu gttsize=120000
options ttm pages_limit=31457280
options ttm page_pool_size=15728640