asxpi/nixOS
asxpi/nixOS
1
0
Fork 0
Code Issues Pull requests Activity

ollama: model store to /mnt/data, gfx1151 rocBLAS workaround, load tuning

Browse source
This commit is contained in:
Sergei Poljanski 2026-07-11 22:36:00 +04:00
commit 22746bcc25
Signed by: asxpi
GPG key ID: 4F8851660FA4121B
1 changed files with 15 additions and 0 deletions
Download patch file Download diff file Expand all files Collapse all files

15
modules/ollama.nix
View file

@ -7,10 +7,25 @@
rocmOverrideGfx = "11.5.1"; # gfx1151, Krackan Point RDNA 3.5
host = "127.0.0.1";
port = 11434;
# Model store on plain-ext4 /mnt/data: model blobs are public data, so
# they don't need to occupy the 512G LUKS root or pay dm-crypt on reads.
# Static user instead of the default DynamicUser — a random dynamic UID
# can't own a persistent store outside /var/lib.
user = "ollama";
group = "ollama";
home = "/mnt/data/ollama";
environmentVariables = {
HSA_OVERRIDE_GFX_VERSION = "11.5.1";
# gfx1151 (RDNA 3.5) lacks tuned hipBLASLt kernels in current ROCm; force
# the stable rocBLAS GEMM path to avoid crashes/garbage output on this iGPU.
ROCBLAS_USE_HIPBLASLT = "0";
OLLAMA_FLASH_ATTENTION = "1";
OLLAMA_KV_CACHE_TYPE = "q8_0";
# Keep the loaded model resident: every eviction re-pays a multi-minute
# 20+ GiB disk read (client keep_alive can still override per-request).
OLLAMA_KEEP_ALIVE = "-1";
# Default 5m load timeout kills large-model loads under memory pressure.
OLLAMA_LOAD_TIMEOUT = "15m";
};
loadModels = [ "qwen3:14b" ];
};