llama-cpp
This commit is contained in:
@@ -11,13 +11,21 @@
|
||||
home = "/media/nas/ssd/nix-app-data/ollama";
|
||||
};
|
||||
|
||||
# services.llama-cpp = {
|
||||
# enable = true;
|
||||
# port = 10400;
|
||||
# host = "0.0.0.0";
|
||||
# openFirewall = true;
|
||||
# model = "/media/nas/ssd/nix-app-data/llama-cpp/models/functionary-small-v3.2-GGUF/functionary-small-v3.2.Q8_0.gguf";
|
||||
# };
|
||||
services.llama-cpp = {
|
||||
enable = true;
|
||||
port = 8127;
|
||||
host = "0.0.0.0";
|
||||
openFirewall = true;
|
||||
model = "/media/nas/ssd/nix-app-data/llama-cpp/models/functionary-small-v3.2-GGUF/functionary-small-v3.2.Q4_0.gguf";
|
||||
extraFlags = [
|
||||
"--n_gpu-layers"
|
||||
"500"
|
||||
"-c"
|
||||
"0"
|
||||
"--numa"
|
||||
"numactl"
|
||||
]
|
||||
};
|
||||
|
||||
services.open-webui = {
|
||||
enable = true;
|
||||
|
||||
Reference in New Issue
Block a user