Commit 214c6c6e9f27

@@ -325,39 +325,7 @@
       endpoint = "${globals.net.vpn.endpoint}";
       endpointPublicKey = "${globals.machines.kerkouane.net.vpn.pubkey}";
     };
-    ollama = {
-      enable = true;
-      # acceleration = "cuda"; # no nvidia :D
-      host = "0.0.0.0"; # Listen on all interfaces for network access
-      port = 11434;
-      loadModels = [
-        # Coding Models
-        "qwen2.5-coder:7b" # Best coding: 88.4% HumanEval, Apache 2.0 (~4-5GB, 10-15 tok/s)
-        "codestral" # Latest coding (Jan 2025): 86.6% HumanEval, #1 LMsys (~14GB, 8-10 tok/s)
 
-        # Reasoning Models
-        "phi4-reasoning" # Best 14B reasoning: outperforms 70B distillation, MIT (~9GB, 6-10 tok/s)
-        "deepseek-r1:7b" # Lightweight reasoning: MIT license (~4.5GB, 8-12 tok/s)
-
-        # Multimodal
-        "qwen2.5vl:7b" # Best vision: beats Llama 3.2 11B, Apache 2.0 (~6GB, 5-8 tok/s)
-
-        # Quick Tasks
-        "phi3.5:3.8b" # Ultra-fast all-rounder: MIT license (~2.4GB, 15-25 tok/s)
-
-        # Tool Calling / OpenCode Support
-        "llama3.1:8b" # Native tool support, good for OpenCode (~4.7GB)
-        "mistral-nemo" # Fast tool calling support (~7.7GB)
-
-        # Legacy (keeping for compatibility)
-        "mistral:7b-instruct-q4_K_M" # General purpose fallback
-      ];
-      environmentVariables = {
-        OLLAMA_MODELS = "/var/lib/ollama/models";
-        OLLAMA_NUM_PARALLEL = "1"; # CPU-only, keep it simple
-        OLLAMA_KEEP_ALIVE = "10m";
-      };
-    };
     smartd = {
       enable = true;
       devices = [ { device = "/dev/nvme0n1"; } ];
@@ -404,49 +372,10 @@
     jayrah
   ];
 
-  # Ollama Prometheus Exporter (Docker-based, built locally)
   systemd.tmpfiles.rules = [
-    "d /var/lib/ollama-exporter 0755 root root -"
     "d /var/lib/git-builds 0755 builder users -"
   ];
 
-  systemd.services.ollama-exporter = {
-    description = "Ollama Prometheus Exporter";
-    after = [
-      "docker.service"
-      "ollama.service"
-    ];
-    requires = [ "docker.service" ];
-    wantedBy = [ "multi-user.target" ];
-
-    serviceConfig = {
-      Type = "simple";
-      Restart = "always";
-      RestartSec = "10s";
-
-      ExecStartPre = [
-        # Stop and remove existing container
-        "-${pkgs.docker}/bin/docker stop ollama-exporter"
-        "-${pkgs.docker}/bin/docker rm ollama-exporter"
-        # Copy source files to build directory (for future manual rebuilds if needed)
-        "${pkgs.coreutils}/bin/cp ${../../tools/ollama-exporter/Dockerfile} /var/lib/ollama-exporter/Dockerfile"
-        "${pkgs.coreutils}/bin/cp ${../../tools/ollama-exporter/ollama_exporter.py} /var/lib/ollama-exporter/ollama_exporter.py"
-        # Build image locally only if it doesn't exist (to avoid DNS timeout issues)
-        "-${pkgs.bash}/bin/bash -c '${pkgs.docker}/bin/docker image inspect ollama-exporter:local >/dev/null 2>&1 || ${pkgs.docker}/bin/docker build -t ollama-exporter:local /var/lib/ollama-exporter'"
-      ];
-
-      ExecStart = ''
-        ${pkgs.docker}/bin/docker run --rm --name ollama-exporter \
-          -p 8000:8000 \
-          -e OLLAMA_HOST=http://localhost:11434 \
-          --network host \
-          ollama-exporter:local
-      '';
-
-      ExecStop = "${pkgs.docker}/bin/docker stop ollama-exporter";
-    };
-  };
-
   # NOTE: NixOS firewall is disabled (see openshift-port-forward.nix).
   # Firewall rules must be added to the nftables config there instead.
   # networking.firewall.allowedTCPPorts = [ 8000 8888 ];