Gentoo Portage Overlays - sci-misc/kvmem-llama-cpp

sci-misc/kvmem-llama-cpp

CUDA llama.cpp runtime with tiered KV memory for long contexts

Screenshots

https://github.com/kvmem/kvmem-llama.cpp

kvmem-llama-cpp-0.17.0

~amd64

cpu_flags_x86_avx cpu_flags_x86_avx2 cpu_flags_x86_bmi2 cpu_flags_x86_f16c cpu_flags_x86_fma3 cpu_flags_x86_sse4_2 +openmp test

View

Download

Browse License: Apache-2.0 MIT

Overlay: ruizhi-overlay

ChangeLog USE Flags Dependencies Reverse Deps Related Bugs

USE Flags

cpu_flags_x86_avx

* This flag is undocumented *

cpu_flags_x86_avx2

* This flag is undocumented *

cpu_flags_x86_bmi2

* This flag is undocumented *

cpu_flags_x86_f16c

* This flag is undocumented *

cpu_flags_x86_fma3

* This flag is undocumented *

cpu_flags_x86_sse4_2

* This flag is undocumented *

+openmp

* This flag is undocumented *

test

Global: Workaround to pull in packages needed to run with FEATURES=test. Portage-2.1.2 handles this internally, so don't set it in make.conf/package.use anymore