See Hiong: Running Llama.cpp on AMD Strix Halo (Radeon 8060S, gfx1151)
blog
Fields
id
sources/seehiong-strix-halo-llama-cppPublisher
See Hiong's Blog
Kind
blog
Accessed
2026-10-11
notes
Reproducible benchmark guide for running llama.cpp with ROCm 7.x on AMD Ryzen AI Max+ 395 (128GB LPDDR5X), documenting TTM pages_limit memory configuration and 7B through 120B benchmarks.
Notes
Reproducible benchmark guide for running llama.cpp with ROCm 7.x on AMD Ryzen AI Max+ 395 (128GB LPDDR5X), documenting TTM pages_limit memory configuration and 7B through 120B benchmarks.
Cross-references
This record references no other records.