diff --git a/configuration.nix b/configuration.nix
index 515fa79..f318840 100644
--- a/configuration.nix
+++ b/configuration.nix
@@ -45,6 +45,8 @@
 
     ./services/soulseek.nix
 
+    ./services/llama-cpp.nix
+
     ./services/ups.nix
 
     ./services/bitwarden.nix
diff --git a/flake.lock b/flake.lock
index c0185ea..7125090 100644
--- a/flake.lock
+++ b/flake.lock
@@ -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 +294,27 @@
         "type": "github"
       }
     },
+    "llamacpp": {
+      "inputs": {
+        "flake-parts": "flake-parts",
+        "nixpkgs": [
+          "nixpkgs"
+        ]
+      },
+      "locked": {
+        "lastModified": 1774806340,
+        "narHash": "sha256-KC0ZkqR8HYHOjQbX3+kxIPv0fsYcjNFYDU9WgZZwNE4=",
+        "owner": "ggml-org",
+        "repo": "llama.cpp",
+        "rev": "7c203670f8d746382247ed369fea7fbf10df8ae0",
+        "type": "github"
+      },
+      "original": {
+        "owner": "ggml-org",
+        "repo": "llama.cpp",
+        "type": "github"
+      }
+    },
     "nix-minecraft": {
       "inputs": {
         "flake-compat": "flake-compat_3",
@@ -330,6 +369,18 @@
         "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 +446,7 @@
         "home-manager": "home-manager",
         "impermanence": "impermanence",
         "lanzaboote": "lanzaboote",
+        "llamacpp": "llamacpp",
         "nix-minecraft": "nix-minecraft",
         "nixos-hardware": "nixos-hardware",
         "nixpkgs": "nixpkgs",
diff --git a/flake.nix b/flake.nix
index e8c4ba9..ec32b6b 100644
--- a/flake.nix
+++ b/flake.nix
@@ -28,6 +28,11 @@
       inputs.nixpkgs.follows = "nixpkgs";
     };
 
+    llamacpp = {
+      url = "github:ggml-org/llama.cpp";
+      inputs.nixpkgs.follows = "nixpkgs";
+    };
+
     srvos = {
       url = "github:nix-community/srvos";
       inputs.nixpkgs.follows = "nixpkgs";
diff --git a/service-configs.nix b/service-configs.nix
index ea35cf5..7ad7830 100644
--- a/service-configs.nix
+++ b/service-configs.nix
@@ -149,6 +149,10 @@ rec {
         port = 5000;
         proto = "tcp";
       };
+      llama_cpp = {
+        port = 6688;
+        proto = "tcp";
+      };
     };
   };
 
diff --git a/services/llama-cpp.nix b/services/llama-cpp.nix
new file mode 100644
index 0000000..de2501c
--- /dev/null
+++ b/services/llama-cpp.nix
@@ -0,0 +1,36 @@
+{
+  pkgs,
+  service_configs,
+  config,
+  inputs,
+  lib,
+  ...
+}:
+{
+  services.llama-cpp = {
+    enable = true;
+    model = builtins.toString (
+      pkgs.fetchurl {
+        url = "https://huggingface.co/mradermacher/Qwen3.5-27B-Claude-4.6-Opus-Reasoning-Distilled-GGUF/resolve/main/Qwen3.5-27B-Claude-4.6-Opus-Reasoning-Distilled.Q5_K_M.gguf";
+        sha256 = "1b08df702dc729104de8894d3f6c6f52505e9f07c5d9b236b3efda3ae187bda2";
+      }
+    );
+    port = service_configs.ports.private.llama_cpp.port;
+    host = "0.0.0.0";
+    package = (lib.optimizePackage inputs.llamacpp.packages.${pkgs.system}.vulkan);
+    extraFlags = [
+      "-ngl"
+      "12"
+      "-c"
+      "16384"
+    ];
+  };
+
+  # have to do this in order to get vulkan to work
+  systemd.services.llama-cpp.serviceConfig.DynamicUser = lib.mkForce false;
+
+  services.caddy.virtualHosts."llm.${service_configs.https.domain}".extraConfig = ''
+    import ${config.age.secrets.caddy_auth.path}
+    reverse_proxy :${builtins.toString config.services.llama-cpp.port}
+  '';
+}