llama-cpp: re-add w/ turboquant
All checks were successful
Build and Deploy / deploy (push) Successful in 28m52s
All checks were successful
Build and Deploy / deploy (push) Successful in 28m52s
This commit is contained in:
@@ -47,6 +47,8 @@
|
||||
|
||||
./services/soulseek.nix
|
||||
|
||||
./services/llama-cpp.nix
|
||||
|
||||
./services/ups.nix
|
||||
./services/monitoring.nix
|
||||
./services/jellyfin-annotations.nix
|
||||
|
||||
53
flake.lock
generated
53
flake.lock
generated
@@ -150,6 +150,24 @@
|
||||
"type": "github"
|
||||
}
|
||||
},
|
||||
"flake-parts": {
|
||||
"inputs": {
|
||||
"nixpkgs-lib": "nixpkgs-lib"
|
||||
},
|
||||
"locked": {
|
||||
"lastModified": 1730504689,
|
||||
"narHash": "sha256-hgmguH29K2fvs9szpq2r3pz2/8cJd2LPS+b4tfNFCwE=",
|
||||
"owner": "hercules-ci",
|
||||
"repo": "flake-parts",
|
||||
"rev": "506278e768c2a08bec68eb62932193e341f55c90",
|
||||
"type": "github"
|
||||
},
|
||||
"original": {
|
||||
"owner": "hercules-ci",
|
||||
"repo": "flake-parts",
|
||||
"type": "github"
|
||||
}
|
||||
},
|
||||
"flake-utils": {
|
||||
"inputs": {
|
||||
"systems": "systems_4"
|
||||
@@ -276,6 +294,28 @@
|
||||
"type": "github"
|
||||
}
|
||||
},
|
||||
"llamacpp": {
|
||||
"inputs": {
|
||||
"flake-parts": "flake-parts",
|
||||
"nixpkgs": [
|
||||
"nixpkgs"
|
||||
]
|
||||
},
|
||||
"locked": {
|
||||
"lastModified": 1775101360,
|
||||
"narHash": "sha256-X1cyWED8lmsGKFc7Pb6nGJ8EVzpPqi5iKcyL8NVVIe8=",
|
||||
"owner": "TheTom",
|
||||
"repo": "llama-cpp-turboquant",
|
||||
"rev": "04eeabb0d344b54ca12d4140b8af8c236ffe7beb",
|
||||
"type": "github"
|
||||
},
|
||||
"original": {
|
||||
"owner": "TheTom",
|
||||
"ref": "feature/turboquant-kv-cache",
|
||||
"repo": "llama-cpp-turboquant",
|
||||
"type": "github"
|
||||
}
|
||||
},
|
||||
"nix-minecraft": {
|
||||
"inputs": {
|
||||
"flake-compat": "flake-compat_3",
|
||||
@@ -330,6 +370,18 @@
|
||||
"type": "github"
|
||||
}
|
||||
},
|
||||
"nixpkgs-lib": {
|
||||
"locked": {
|
||||
"lastModified": 1730504152,
|
||||
"narHash": "sha256-lXvH/vOfb4aGYyvFmZK/HlsNsr/0CVWlwYvo2rxJk3s=",
|
||||
"type": "tarball",
|
||||
"url": "https://github.com/NixOS/nixpkgs/archive/cc2f28000298e1269cea6612cd06ec9979dd5d7f.tar.gz"
|
||||
},
|
||||
"original": {
|
||||
"type": "tarball",
|
||||
"url": "https://github.com/NixOS/nixpkgs/archive/cc2f28000298e1269cea6612cd06ec9979dd5d7f.tar.gz"
|
||||
}
|
||||
},
|
||||
"nixpkgs-p2pool-module": {
|
||||
"flake": false,
|
||||
"locked": {
|
||||
@@ -395,6 +447,7 @@
|
||||
"home-manager": "home-manager",
|
||||
"impermanence": "impermanence",
|
||||
"lanzaboote": "lanzaboote",
|
||||
"llamacpp": "llamacpp",
|
||||
"nix-minecraft": "nix-minecraft",
|
||||
"nixos-hardware": "nixos-hardware",
|
||||
"nixpkgs": "nixpkgs",
|
||||
|
||||
@@ -28,6 +28,11 @@
|
||||
inputs.nixpkgs.follows = "nixpkgs";
|
||||
};
|
||||
|
||||
llamacpp = {
|
||||
url = "github:TheTom/llama-cpp-turboquant/feature/turboquant-kv-cache";
|
||||
inputs.nixpkgs.follows = "nixpkgs";
|
||||
};
|
||||
|
||||
srvos = {
|
||||
url = "github:nix-community/srvos";
|
||||
inputs.nixpkgs.follows = "nixpkgs";
|
||||
|
||||
@@ -169,6 +169,10 @@ rec {
|
||||
port = 9162;
|
||||
proto = "tcp";
|
||||
};
|
||||
llama_cpp = {
|
||||
port = 6688;
|
||||
proto = "tcp";
|
||||
};
|
||||
};
|
||||
};
|
||||
|
||||
|
||||
40
services/llama-cpp.nix
Normal file
40
services/llama-cpp.nix
Normal file
@@ -0,0 +1,40 @@
|
||||
{
|
||||
pkgs,
|
||||
service_configs,
|
||||
config,
|
||||
inputs,
|
||||
lib,
|
||||
...
|
||||
}:
|
||||
{
|
||||
services.llama-cpp = {
|
||||
enable = true;
|
||||
model = toString (
|
||||
pkgs.fetchurl {
|
||||
url = "https://huggingface.co/Jackrong/Qwen3.5-9B-Claude-4.6-Opus-Reasoning-Distilled-v2-GGUF/resolve/main/Qwen3.5-9B.Q4_K_M.gguf";
|
||||
sha256 = "8fbbc7b04a7d4b052d14b7aa97c8bf2014d39ceca8c2baaa043711712ba71ccc";
|
||||
}
|
||||
);
|
||||
port = service_configs.ports.private.llama_cpp.port;
|
||||
host = "0.0.0.0";
|
||||
package = (lib.optimizePackage inputs.llamacpp.packages.${pkgs.system}.vulkan);
|
||||
extraFlags = [
|
||||
"-ngl"
|
||||
"12"
|
||||
"-c"
|
||||
"16384"
|
||||
"-ctk"
|
||||
"q8_0"
|
||||
"-ctv"
|
||||
"turbo3"
|
||||
];
|
||||
};
|
||||
|
||||
# have to do this in order to get vulkan to work
|
||||
systemd.services.llama-cpp.serviceConfig.DynamicUser = lib.mkForce false;
|
||||
|
||||
services.caddy.virtualHosts."llm.${service_configs.https.domain}".extraConfig = ''
|
||||
import ${config.age.secrets.caddy_auth.path}
|
||||
reverse_proxy :${toString config.services.llama-cpp.port}
|
||||
'';
|
||||
}
|
||||
Reference in New Issue
Block a user