migrate model cards to .toml files

dashboard: add peer: true to package lock (#1162 )
this happens every time i run npm install - lets upstream it ## testing dashboard builds and renders
2026-01-15 17:39:45 -05:00 · 2026-01-15 17:07:48 +00:00 · 2026-01-15 17:01:43 +00:00 · 2026-01-15 16:51:46 +00:00 · 2026-01-15 13:21:58 +00:00 · 2026-01-14 16:33:17 +00:00
52 changed files with 2020 additions and 1067 deletions
--- a/.github/workflows/build-app.yml
+++ b/.github/workflows/build-app.yml
@@ -113,11 +113,22 @@ jobs:
          uv python install
          uv sync --locked

+      - name: Install Nix
+        uses: cachix/install-nix-action@v31
+        with:
+          nix_path: nixpkgs=channel:nixos-unstable
+
+      - name: Configure Cachix
+        uses: cachix/cachix-action@v14
+        with:
+          name: exo
+          authToken: "${{ secrets.CACHIX_AUTH_TOKEN }}"
+
      - name: Build dashboard
        run: |
-          cd dashboard
-          npm ci
-          npm run build
+          DASHBOARD_OUT=$(nix build .#dashboard --print-build-logs --no-link --print-out-paths)
+          mkdir -p dashboard/build
+          cp -r "$DASHBOARD_OUT"/* dashboard/build/

      - name: Install Sparkle CLI
        run: |
--- a/Cargo.lock
+++ b/Cargo.lock
@@ -4340,25 +4340,6 @@ dependencies = [
 "libc",
 ]

-[[package]]
-name = "system_custodian"
-version = "0.0.1"
-dependencies = [
- "delegate",
- "derive_more",
- "either",
- "extend",
- "futures",
- "futures-timer",
- "impl-trait-for-tuples",
- "keccak-const",
- "log",
- "thiserror 2.0.17",
- "tokio",
- "tracing-subscriber",
- "util",
-]
-
 [[package]]
 name = "tagptr"
 version = "0.2.0"
--- a/Cargo.toml
+++ b/Cargo.toml
@@ -3,7 +3,6 @@ resolver = "3"
 members = [
    "rust/networking",
    "rust/exo_pyo3_bindings",
-    "rust/system_custodian",
    "rust/util",
 ]

@@ -25,7 +24,6 @@ opt-level = 3
 [workspace.dependencies]
 ## Crate members as common dependencies
 networking = { path = "rust/networking" }
-system_custodian = { path = "rust/system_custodian" }
 util = { path = "rust/util" }

 # Proc-macro authoring tools
--- a/dashboard/dashboard.nix
+++ b/dashboard/dashboard.nix
@@ -0,0 +1,60 @@
+{ lib
+, config
+, dream2nix
+, ...
+}:
+let
+  # Read and parse the lock file
+  rawLockFile = builtins.fromJSON (builtins.readFile "${config.deps.dashboardSrc}/package-lock.json");
+
+  # For packages with bundleDependencies, filter out deps that are bundled
+  # (bundled deps are inside the tarball, not separate lockfile entries)
+  fixedPackages = lib.mapAttrs
+    (path: entry:
+      if entry ? bundleDependencies && entry.bundleDependencies != [ ]
+      then entry // {
+        dependencies = lib.filterAttrs
+          (name: _: !(lib.elem name entry.bundleDependencies))
+          (entry.dependencies or { });
+      }
+      else entry
+    )
+    (rawLockFile.packages or { });
+
+  fixedLockFile = rawLockFile // { packages = fixedPackages; };
+in
+{
+  imports = [
+    dream2nix.modules.dream2nix.nodejs-package-lock-v3
+    dream2nix.modules.dream2nix.nodejs-granular-v3
+  ];
+
+  name = "exo-dashboard";
+  version = "1.0.0";
+
+  mkDerivation = {
+    src = config.deps.dashboardSrc;
+
+    buildPhase = ''
+      runHook preBuild
+      npm run build
+      runHook postBuild
+    '';
+
+    installPhase = ''
+      runHook preInstall
+      cp -r build $out/build
+      runHook postInstall
+    '';
+  };
+
+  deps = { nixpkgs, ... }: {
+    inherit (nixpkgs) stdenv;
+    dashboardSrc = null; # Injected by parts.nix
+  };
+
+  nodejs-package-lock-v3 = {
+    # Don't use packageLockFile - provide the fixed lock content directly
+    packageLock = fixedLockFile;
+  };
+}
--- a/dashboard/package-lock.json
+++ b/dashboard/package-lock.json
@@ -863,6 +863,7 @@
 			"integrity": "sha512-oH8tXw7EZnie8FdOWYrF7Yn4IKrqTFHhXvl8YxXxbKwTMcD/5NNCryUSEXRk2ZR4ojnub0P8rNrsVGHXWqIDtA==",
 			"dev": true,
 			"license": "MIT",
+			"peer": true,
 			"dependencies": {
 				"@standard-schema/spec": "^1.0.0",
 				"@sveltejs/acorn-typescript": "^1.0.5",
@@ -902,6 +903,7 @@
 			"integrity": "sha512-Y1Cs7hhTc+a5E9Va/xwKlAJoariQyHY+5zBgCZg4PFWNYQ1nMN9sjK1zhw1gK69DuqVP++sht/1GZg1aRwmAXQ==",
 			"dev": true,
 			"license": "MIT",
+			"peer": true,
 			"dependencies": {
 				"@sveltejs/vite-plugin-svelte-inspector": "^4.0.1",
 				"debug": "^4.4.1",
@@ -1518,6 +1520,7 @@
 			"integrity": "sha512-LCCV0HdSZZZb34qifBsyWlUmok6W7ouER+oQIGBScS8EsZsQbrtFTUrDX4hOl+CS6p7cnNC4td+qrSVGSCTUfQ==",
 			"dev": true,
 			"license": "MIT",
+			"peer": true,
 			"dependencies": {
 				"undici-types": "~6.21.0"
 			}
@@ -1527,6 +1530,7 @@
 			"resolved": "https://registry.npmjs.org/acorn/-/acorn-8.15.0.tgz",
 			"integrity": "sha512-NZyJarBfL7nWwIq+FDL6Zp/yHEhePMNnnJ0y3qfieCrmNvYct8uvtiV41UvlSe6apAfk0fY1FbWx+NwfmpvtTg==",
 			"license": "MIT",
+			"peer": true,
 			"bin": {
 				"acorn": "bin/acorn"
 			},
@@ -1939,6 +1943,7 @@
 			"integrity": "sha512-fmTRWbNMmsmWq6xJV8D19U/gw/bwrHfNXxrIN+HfZgnzqTHp9jOmKMhsTUjXOJnZOdZY9Q28y4yebKzqDKlxlQ==",
 			"dev": true,
 			"license": "ISC",
+			"peer": true,
 			"engines": {
 				"node": ">=12"
 			}
@@ -2646,6 +2651,7 @@
 			"integrity": "sha512-5gTmgEY/sqK6gFXLIsQNH19lWb4ebPDLA4SdLP7dsWkIXHWlG66oPuVvXSGFPppYZz8ZDZq0dYYrbHfBCVUb1Q==",
 			"dev": true,
 			"license": "MIT",
+			"peer": true,
 			"engines": {
 				"node": ">=12"
 			},
@@ -2833,6 +2839,7 @@
 			"resolved": "https://registry.npmjs.org/svelte/-/svelte-5.45.3.tgz",
 			"integrity": "sha512-ngKXNhNvwPzF43QqEhDOue7TQTrG09em1sd4HBxVF0Wr2gopAmdEWan+rgbdgK4fhBtSOTJO8bYU4chUG7VXZQ==",
 			"license": "MIT",
+			"peer": true,
 			"dependencies": {
 				"@jridgewell/remapping": "^2.3.4",
 				"@jridgewell/sourcemap-codec": "^1.5.0",
@@ -2977,6 +2984,7 @@
 			"integrity": "sha512-jl1vZzPDinLr9eUt3J/t7V6FgNEw9QjvBPdysz9KfQDD41fQrC2Y4vKQdiaUpFT4bXlb1RHhLpp8wtm6M5TgSw==",
 			"dev": true,
 			"license": "Apache-2.0",
+			"peer": true,
 			"bin": {
 				"tsc": "bin/tsc",
 				"tsserver": "bin/tsserver"
@@ -2998,6 +3006,7 @@
 			"integrity": "sha512-+Oxm7q9hDoLMyJOYfUYBuHQo+dkAloi33apOPP56pzj+vsdJDzr+j1NISE5pyaAuKL4A3UD34qd0lx5+kfKp2g==",
 			"dev": true,
 			"license": "MIT",
+			"peer": true,
 			"dependencies": {
 				"esbuild": "^0.25.0",
 				"fdir": "^6.4.4",
--- a/dashboard/parts.nix
+++ b/dashboard/parts.nix
@@ -0,0 +1,44 @@
+{ inputs, ... }:
+{
+  perSystem =
+    { pkgs, lib, ... }:
+    let
+      # Filter source to only include dashboard directory
+      src = lib.cleanSourceWith {
+        src = inputs.self;
+        filter =
+          path: type:
+          let
+            baseName = builtins.baseNameOf path;
+            inDashboardDir =
+              (lib.hasInfix "/dashboard/" path)
+              || (lib.hasSuffix "/dashboard" (builtins.dirOf path))
+              || (baseName == "dashboard" && type == "directory");
+          in
+          inDashboardDir;
+      };
+
+      # Build the dashboard with dream2nix (includes node_modules in output)
+      dashboardFull = inputs.dream2nix.lib.evalModules {
+        packageSets.nixpkgs = pkgs;
+        modules = [
+          ./dashboard.nix
+          {
+            paths.projectRoot = inputs.self;
+            paths.projectRootFile = "flake.nix";
+            paths.package = inputs.self + "/dashboard";
+          }
+          # Inject the filtered source
+          {
+            deps.dashboardSrc = lib.mkForce "${src}/dashboard";
+          }
+        ];
+      };
+    in
+    {
+      # Extract just the static site from the full build
+      packages.dashboard = pkgs.runCommand "exo-dashboard" { } ''
+        cp -r ${dashboardFull}/build $out
+      '';
+    };
+}
--- a/flake.lock
+++ b/flake.lock
@@ -1,5 +1,42 @@
 {
  "nodes": {
+    "crane": {
+      "locked": {
+        "lastModified": 1767744144,
+        "narHash": "sha256-9/9ntI0D+HbN4G0TrK3KmHbTvwgswz7p8IEJsWyef8Q=",
+        "owner": "ipetkov",
+        "repo": "crane",
+        "rev": "2fb033290bf6b23f226d4c8b32f7f7a16b043d7e",
+        "type": "github"
+      },
+      "original": {
+        "owner": "ipetkov",
+        "repo": "crane",
+        "type": "github"
+      }
+    },
+    "dream2nix": {
+      "inputs": {
+        "nixpkgs": [
+          "nixpkgs"
+        ],
+        "purescript-overlay": "purescript-overlay",
+        "pyproject-nix": "pyproject-nix"
+      },
+      "locked": {
+        "lastModified": 1765953015,
+        "narHash": "sha256-5FBZbbWR1Csp3Y2icfRkxMJw/a/5FGg8hCXej2//bbI=",
+        "owner": "nix-community",
+        "repo": "dream2nix",
+        "rev": "69eb01fa0995e1e90add49d8ca5bcba213b0416f",
+        "type": "github"
+      },
+      "original": {
+        "owner": "nix-community",
+        "repo": "dream2nix",
+        "type": "github"
+      }
+    },
    "fenix": {
      "inputs": {
        "nixpkgs": [
@@ -8,11 +45,11 @@
        "rust-analyzer-src": "rust-analyzer-src"
      },
      "locked": {
-        "lastModified": 1761893049,
-        "narHash": "sha256-1TtFDPhC+ZsrOOtBnry1EZC+WipTTvsOVjIEVugqji8=",
+        "lastModified": 1768287139,
+        "narHash": "sha256-nsXFt0OzUi6K7dUzzJD5/v9e0Ic+fvclfIW936/43ZM=",
        "owner": "nix-community",
        "repo": "fenix",
-        "rev": "c2ac9a5c0d6d16630c3b225b874bd14528d1abe6",
+        "rev": "a4a3aa956931f90f35453cb519e4545e9ad7f773",
        "type": "github"
      },
      "original": {
@@ -21,6 +58,22 @@
        "type": "github"
      }
    },
+    "flake-compat": {
+      "flake": false,
+      "locked": {
+        "lastModified": 1696426674,
+        "narHash": "sha256-kvjfFW7WAETZlt09AgDn1MrtKzP7t90Vf7vypd3OL1U=",
+        "owner": "edolstra",
+        "repo": "flake-compat",
+        "rev": "0f9255e01c2351cc7d116c072cb317785dd33b33",
+        "type": "github"
+      },
+      "original": {
+        "owner": "edolstra",
+        "repo": "flake-compat",
+        "type": "github"
+      }
+    },
    "flake-parts": {
      "inputs": {
        "nixpkgs-lib": [
@@ -43,11 +96,11 @@
    },
    "nixpkgs": {
      "locked": {
-        "lastModified": 1761672384,
-        "narHash": "sha256-o9KF3DJL7g7iYMZq9SWgfS1BFlNbsm6xplRjVlOCkXI=",
+        "lastModified": 1768127708,
+        "narHash": "sha256-1Sm77VfZh3mU0F5OqKABNLWxOuDeHIlcFjsXeeiPazs=",
        "owner": "NixOS",
        "repo": "nixpkgs",
-        "rev": "08dacfca559e1d7da38f3cf05f1f45ee9bfd213c",
+        "rev": "ffbc9f8cbaacfb331b6017d5a5abb21a492c9a38",
        "type": "github"
      },
      "original": {
@@ -57,22 +110,85 @@
        "type": "github"
      }
    },
+    "nixpkgs-swift": {
+      "locked": {
+        "lastModified": 1761672384,
+        "narHash": "sha256-o9KF3DJL7g7iYMZq9SWgfS1BFlNbsm6xplRjVlOCkXI=",
+        "owner": "NixOS",
+        "repo": "nixpkgs",
+        "rev": "08dacfca559e1d7da38f3cf05f1f45ee9bfd213c",
+        "type": "github"
+      },
+      "original": {
+        "owner": "NixOS",
+        "repo": "nixpkgs",
+        "rev": "08dacfca559e1d7da38f3cf05f1f45ee9bfd213c",
+        "type": "github"
+      }
+    },
+    "purescript-overlay": {
+      "inputs": {
+        "flake-compat": "flake-compat",
+        "nixpkgs": [
+          "dream2nix",
+          "nixpkgs"
+        ],
+        "slimlock": "slimlock"
+      },
+      "locked": {
+        "lastModified": 1728546539,
+        "narHash": "sha256-Sws7w0tlnjD+Bjck1nv29NjC5DbL6nH5auL9Ex9Iz2A=",
+        "owner": "thomashoneyman",
+        "repo": "purescript-overlay",
+        "rev": "4ad4c15d07bd899d7346b331f377606631eb0ee4",
+        "type": "github"
+      },
+      "original": {
+        "owner": "thomashoneyman",
+        "repo": "purescript-overlay",
+        "type": "github"
+      }
+    },
+    "pyproject-nix": {
+      "inputs": {
+        "nixpkgs": [
+          "dream2nix",
+          "nixpkgs"
+        ]
+      },
+      "locked": {
+        "lastModified": 1763017646,
+        "narHash": "sha256-Z+R2lveIp6Skn1VPH3taQIuMhABg1IizJd8oVdmdHsQ=",
+        "owner": "pyproject-nix",
+        "repo": "pyproject.nix",
+        "rev": "47bd6f296502842643078d66128f7b5e5370790c",
+        "type": "github"
+      },
+      "original": {
+        "owner": "pyproject-nix",
+        "repo": "pyproject.nix",
+        "type": "github"
+      }
+    },
    "root": {
      "inputs": {
+        "crane": "crane",
+        "dream2nix": "dream2nix",
        "fenix": "fenix",
        "flake-parts": "flake-parts",
        "nixpkgs": "nixpkgs",
+        "nixpkgs-swift": "nixpkgs-swift",
        "treefmt-nix": "treefmt-nix"
      }
    },
    "rust-analyzer-src": {
      "flake": false,
      "locked": {
-        "lastModified": 1761849405,
-        "narHash": "sha256-igXdvC+WCUN+3gnfk+ptT7rMmxQuY6WbIg1rXMUN1DM=",
+        "lastModified": 1768224240,
+        "narHash": "sha256-Pp1dDrXKPBUJReZnnDElFyHYn67XTd48zRhToheLjtk=",
        "owner": "rust-lang",
        "repo": "rust-analyzer",
-        "rev": "f7de8ae045a5fe80f1203c5a1c3015b05f7c3550",
+        "rev": "725349602e525df37f377701e001fe8aab807878",
        "type": "github"
      },
      "original": {
@@ -82,6 +198,28 @@
        "type": "github"
      }
    },
+    "slimlock": {
+      "inputs": {
+        "nixpkgs": [
+          "dream2nix",
+          "purescript-overlay",
+          "nixpkgs"
+        ]
+      },
+      "locked": {
+        "lastModified": 1688756706,
+        "narHash": "sha256-xzkkMv3neJJJ89zo3o2ojp7nFeaZc2G0fYwNXNJRFlo=",
+        "owner": "thomashoneyman",
+        "repo": "slimlock",
+        "rev": "cf72723f59e2340d24881fd7bf61cb113b4c407c",
+        "type": "github"
+      },
+      "original": {
+        "owner": "thomashoneyman",
+        "repo": "slimlock",
+        "type": "github"
+      }
+    },
    "treefmt-nix": {
      "inputs": {
        "nixpkgs": [
@@ -89,11 +227,11 @@
        ]
      },
      "locked": {
-        "lastModified": 1762938485,
-        "narHash": "sha256-AlEObg0syDl+Spi4LsZIBrjw+snSVU4T8MOeuZJUJjM=",
+        "lastModified": 1768158989,
+        "narHash": "sha256-67vyT1+xClLldnumAzCTBvU0jLZ1YBcf4vANRWP3+Ak=",
        "owner": "numtide",
        "repo": "treefmt-nix",
-        "rev": "5b4ee75aeefd1e2d5a1cc43cf6ba65eba75e83e4",
+        "rev": "e96d59dff5c0d7fddb9d113ba108f03c3ef99eca",
        "type": "github"
      },
      "original": {
--- a/flake.nix
+++ b/flake.nix
@@ -9,6 +9,8 @@
      inputs.nixpkgs-lib.follows = "nixpkgs";
    };

+    crane.url = "github:ipetkov/crane";
+
    fenix = {
      url = "github:nix-community/fenix";
      inputs.nixpkgs.follows = "nixpkgs";
@@ -18,6 +20,14 @@
      url = "github:numtide/treefmt-nix";
      inputs.nixpkgs.follows = "nixpkgs";
    };
+
+    dream2nix = {
+      url = "github:nix-community/dream2nix";
+      inputs.nixpkgs.follows = "nixpkgs";
+    };
+
+    # Pinned nixpkgs for swift-format (swift is broken on x86_64-linux in newer nixpkgs)
+    nixpkgs-swift.url = "github:NixOS/nixpkgs/08dacfca559e1d7da38f3cf05f1f45ee9bfd213c";
  };

  nixConfig = {
@@ -36,12 +46,16 @@

      imports = [
        inputs.treefmt-nix.flakeModule
+        ./dashboard/parts.nix
+        ./rust/parts.nix
      ];

      perSystem =
-        { config, inputs', pkgs, lib, ... }:
+        { config, self', inputs', pkgs, lib, system, ... }:
        let
          fenixToolchain = inputs'.fenix.packages.complete;
+          # Use pinned nixpkgs for swift-format (swift is broken on x86_64-linux in newer nixpkgs)
+          pkgsSwift = import inputs.nixpkgs-swift { inherit system; };
        in
        {
          treefmt = {
@@ -54,13 +68,16 @@
              };
              rustfmt = {
                enable = true;
-                package = fenixToolchain.rustfmt;
+                package = config.rust.toolchain;
              };
              prettier = {
                enable = true;
                includes = [ "*.ts" ];
              };
-              swift-format.enable = true;
+              swift-format = {
+                enable = true;
+                package = pkgsSwift.swiftPackages.swift-format;
+              };
            };
          };

@@ -71,6 +88,8 @@
          '';

          devShells.default = with pkgs; pkgs.mkShell {
+            inputsFrom = [ self'.checks.cargo-build ];
+
            packages =
              [
                # FORMATTING
@@ -83,14 +102,8 @@
                basedpyright

                # RUST
-                (fenixToolchain.withComponents [
-                  "cargo"
-                  "rustc"
-                  "clippy"
-                  "rustfmt"
-                  "rust-src"
-                ])
-                rustup # Just here to make RustRover happy
+                config.rust.toolchain
+                maturin

                # NIX
                nixpkgs-fmt
@@ -102,30 +115,20 @@
                just
                jq
              ]
-              ++ (pkgs.lib.optionals pkgs.stdenv.isLinux [
-                # IFCONFIG
+              ++ lib.optionals stdenv.isLinux [
                unixtools.ifconfig
-
-                # Build dependencies for Linux
-                pkg-config
-                openssl
-              ])
-              ++ (pkgs.lib.optionals pkgs.stdenv.isDarwin [
-                # MACMON
+              ]
+              ++ lib.optionals stdenv.isDarwin [
                macmon
-              ]);
+              ];
+
+            OPENSSL_NO_VENDOR = "1";

            shellHook = ''
-              # PYTHON
-              export LD_LIBRARY_PATH="$LD_LIBRARY_PATH:${pkgs.python313}/lib"
-              ${lib.optionalString pkgs.stdenv.isLinux ''
-                # Build environment for Linux
-                export PKG_CONFIG_PATH="${pkgs.openssl.dev}/lib/pkgconfig:$PKG_CONFIG_PATH"
-                export LD_LIBRARY_PATH="${pkgs.openssl.out}/lib:$LD_LIBRARY_PATH"
+              export LD_LIBRARY_PATH="$LD_LIBRARY_PATH:${python313}/lib"
+              ${lib.optionalString stdenv.isLinux ''
+                export LD_LIBRARY_PATH="${openssl.out}/lib:$LD_LIBRARY_PATH"
              ''}
-              echo
-              echo "🍎🍎 Run 'just <recipe>' to get started"
-              just --list
            '';
          };
        };
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -23,6 +23,7 @@ dependencies = [
    "tiktoken>=0.12.0", # required for kimi k2 tokenizer
    "hypercorn>=0.18.0",
    "openai-harmony>=0.0.8",
+    "tomlkit>=0.14.0",
 ]

 [project.scripts]
--- a/resources/model_cards/deepseek-v3.1-4bit.toml
+++ b/resources/model_cards/deepseek-v3.1-4bit.toml
@@ -0,0 +1,15 @@
+short_id = "deepseek-v3.1-4bit"
+model_id = "mlx-community/DeepSeek-V3.1-4bit"
+name = "DeepSeek V3.1 (4-bit)"
+description = "DeepSeek V3.1 is a large language model trained on the DeepSeek V3.1 dataset."
+tags = []
+
+[metadata]
+model_id = "mlx-community/DeepSeek-V3.1-4bit"
+pretty_name = "DeepSeek V3.1 (4-bit)"
+n_layers = 61
+hidden_size = 7168
+supports_tensor = true
+
+[metadata.storage_size]
+in_bytes = 405874409472
--- a/resources/model_cards/deepseek-v3.1-8bit.toml
+++ b/resources/model_cards/deepseek-v3.1-8bit.toml
@@ -0,0 +1,15 @@
+short_id = "deepseek-v3.1-8bit"
+model_id = "mlx-community/DeepSeek-V3.1-8bit"
+name = "DeepSeek V3.1 (8-bit)"
+description = "DeepSeek V3.1 is a large language model trained on the DeepSeek V3.1 dataset."
+tags = []
+
+[metadata]
+model_id = "mlx-community/DeepSeek-V3.1-8bit"
+pretty_name = "DeepSeek V3.1 (8-bit)"
+n_layers = 61
+hidden_size = 7168
+supports_tensor = true
+
+[metadata.storage_size]
+in_bytes = 765577920512
--- a/resources/model_cards/glm-4.5-air-8bit.toml
+++ b/resources/model_cards/glm-4.5-air-8bit.toml
@@ -0,0 +1,15 @@
+short_id = "glm-4.5-air-8bit"
+model_id = "mlx-community/GLM-4.5-Air-8bit"
+name = "GLM 4.5 Air 8bit"
+description = "GLM 4.5 Air 8bit"
+tags = []
+
+[metadata]
+model_id = "mlx-community/GLM-4.5-Air-8bit"
+pretty_name = "GLM 4.5 Air 8bit"
+n_layers = 46
+hidden_size = 4096
+supports_tensor = false
+
+[metadata.storage_size]
+in_bytes = 122406567936
--- a/resources/model_cards/glm-4.5-air-bf16.toml
+++ b/resources/model_cards/glm-4.5-air-bf16.toml
@@ -0,0 +1,15 @@
+short_id = "glm-4.5-air-bf16"
+model_id = "mlx-community/GLM-4.5-Air-bf16"
+name = "GLM 4.5 Air bf16"
+description = "GLM 4.5 Air bf16"
+tags = []
+
+[metadata]
+model_id = "mlx-community/GLM-4.5-Air-bf16"
+pretty_name = "GLM 4.5 Air bf16"
+n_layers = 46
+hidden_size = 4096
+supports_tensor = true
+
+[metadata.storage_size]
+in_bytes = 229780750336
--- a/resources/model_cards/glm-4.7-4bit.toml
+++ b/resources/model_cards/glm-4.7-4bit.toml
@@ -0,0 +1,15 @@
+short_id = "glm-4.7-4bit"
+model_id = "mlx-community/GLM-4.7-4bit"
+name = "GLM 4.7 4bit"
+description = "GLM 4.7 4bit"
+tags = []
+
+[metadata]
+model_id = "mlx-community/GLM-4.7-4bit"
+pretty_name = "GLM 4.7 4bit"
+n_layers = 91
+hidden_size = 5120
+supports_tensor = true
+
+[metadata.storage_size]
+in_bytes = 198556925568
--- a/resources/model_cards/glm-4.7-6bit.toml
+++ b/resources/model_cards/glm-4.7-6bit.toml
@@ -0,0 +1,15 @@
+short_id = "glm-4.7-6bit"
+model_id = "mlx-community/GLM-4.7-6bit"
+name = "GLM 4.7 6bit"
+description = "GLM 4.7 6bit"
+tags = []
+
+[metadata]
+model_id = "mlx-community/GLM-4.7-6bit"
+pretty_name = "GLM 4.7 6bit"
+n_layers = 91
+hidden_size = 5120
+supports_tensor = true
+
+[metadata.storage_size]
+in_bytes = 286737579648
--- a/resources/model_cards/glm-4.7-8bit-gs32.toml
+++ b/resources/model_cards/glm-4.7-8bit-gs32.toml
@@ -0,0 +1,15 @@
+short_id = "glm-4.7-8bit-gs32"
+model_id = "mlx-community/GLM-4.7-8bit-gs32"
+name = "GLM 4.7 8bit (gs32)"
+description = "GLM 4.7 8bit (gs32)"
+tags = []
+
+[metadata]
+model_id = "mlx-community/GLM-4.7-8bit-gs32"
+pretty_name = "GLM 4.7 8bit (gs32)"
+n_layers = 91
+hidden_size = 5120
+supports_tensor = true
+
+[metadata.storage_size]
+in_bytes = 396963397248
--- a/resources/model_cards/gpt-oss-120b-MXFP4-Q8.toml
+++ b/resources/model_cards/gpt-oss-120b-MXFP4-Q8.toml
@@ -0,0 +1,15 @@
+short_id = "gpt-oss-120b-MXFP4-Q8"
+model_id = "mlx-community/gpt-oss-120b-MXFP4-Q8"
+name = "GPT-OSS 120B (MXFP4-Q8, MLX)"
+description = "OpenAI's GPT-OSS 120B is a 117B-parameter Mixture-of-Experts model designed for high-reasoning and general-purpose use; this variant is a 4-bit MLX conversion for Apple Silicon."
+tags = []
+
+[metadata]
+model_id = "mlx-community/gpt-oss-120b-MXFP4-Q8"
+pretty_name = "GPT-OSS 120B (MXFP4-Q8, MLX)"
+n_layers = 36
+hidden_size = 2880
+supports_tensor = true
+
+[metadata.storage_size]
+in_bytes = 70652212224
--- a/resources/model_cards/gpt-oss-20b-4bit.toml
+++ b/resources/model_cards/gpt-oss-20b-4bit.toml
@@ -0,0 +1,15 @@
+short_id = "gpt-oss-20b-4bit"
+model_id = "mlx-community/gpt-oss-20b-MXFP4-Q4"
+name = "GPT-OSS 20B (MXFP4-Q4, MLX)"
+description = "OpenAI's GPT-OSS 20B is a medium-sized MoE model for lower-latency and local or specialized use cases; this MLX variant uses MXFP4 4-bit quantization."
+tags = []
+
+[metadata]
+model_id = "mlx-community/gpt-oss-20b-MXFP4-Q4"
+pretty_name = "GPT-OSS 20B (MXFP4-Q4, MLX)"
+n_layers = 24
+hidden_size = 2880
+supports_tensor = true
+
+[metadata.storage_size]
+in_bytes = 12025908224
--- a/resources/model_cards/kimi-k2-instruct-4bit.toml
+++ b/resources/model_cards/kimi-k2-instruct-4bit.toml
@@ -0,0 +1,15 @@
+short_id = "kimi-k2-instruct-4bit"
+model_id = "mlx-community/Kimi-K2-Instruct-4bit"
+name = "Kimi K2 Instruct (4-bit)"
+description = "Kimi K2 is a large language model trained on the Kimi K2 dataset."
+tags = []
+
+[metadata]
+model_id = "mlx-community/Kimi-K2-Instruct-4bit"
+pretty_name = "Kimi K2 Instruct (4-bit)"
+n_layers = 61
+hidden_size = 7168
+supports_tensor = true
+
+[metadata.storage_size]
+in_bytes = 620622774272
--- a/resources/model_cards/kimi-k2-thinking.toml
+++ b/resources/model_cards/kimi-k2-thinking.toml
@@ -0,0 +1,15 @@
+short_id = "kimi-k2-thinking"
+model_id = "mlx-community/Kimi-K2-Thinking"
+name = "Kimi K2 Thinking (4-bit)"
+description = "Kimi K2 Thinking is the latest, most capable version of open-source thinking model."
+tags = []
+
+[metadata]
+model_id = "mlx-community/Kimi-K2-Thinking"
+pretty_name = "Kimi K2 Thinking (4-bit)"
+n_layers = 61
+hidden_size = 7168
+supports_tensor = true
+
+[metadata.storage_size]
+in_bytes = 706522120192
--- a/resources/model_cards/llama-3.1-70b.toml
+++ b/resources/model_cards/llama-3.1-70b.toml
@@ -0,0 +1,15 @@
+short_id = "llama-3.1-70b"
+model_id = "mlx-community/Meta-Llama-3.1-70B-Instruct-4bit"
+name = "Llama 3.1 70B (4-bit)"
+description = "Llama 3.1 is a large language model trained on the Llama 3.1 dataset."
+tags = []
+
+[metadata]
+model_id = "mlx-community/Meta-Llama-3.1-70B-Instruct-4bit"
+pretty_name = "Llama 3.1 70B (4-bit)"
+n_layers = 80
+hidden_size = 8192
+supports_tensor = true
+
+[metadata.storage_size]
+in_bytes = 40652242944
--- a/resources/model_cards/llama-3.1-8b-8bit.toml
+++ b/resources/model_cards/llama-3.1-8b-8bit.toml
@@ -0,0 +1,15 @@
+short_id = "llama-3.1-8b-8bit"
+model_id = "mlx-community/Meta-Llama-3.1-8B-Instruct-8bit"
+name = "Llama 3.1 8B (8-bit)"
+description = "Llama 3.1 is a large language model trained on the Llama 3.1 dataset."
+tags = []
+
+[metadata]
+model_id = "mlx-community/Meta-Llama-3.1-8B-Instruct-8bit"
+pretty_name = "Llama 3.1 8B (8-bit)"
+n_layers = 32
+hidden_size = 4096
+supports_tensor = true
+
+[metadata.storage_size]
+in_bytes = 8954839040
--- a/resources/model_cards/llama-3.1-8b-bf16.toml
+++ b/resources/model_cards/llama-3.1-8b-bf16.toml
@@ -0,0 +1,15 @@
+short_id = "llama-3.1-8b-bf16"
+model_id = "mlx-community/Meta-Llama-3.1-8B-Instruct-bf16"
+name = "Llama 3.1 8B (BF16)"
+description = "Llama 3.1 is a large language model trained on the Llama 3.1 dataset."
+tags = []
+
+[metadata]
+model_id = "mlx-community/Meta-Llama-3.1-8B-Instruct-bf16"
+pretty_name = "Llama 3.1 8B (BF16)"
+n_layers = 32
+hidden_size = 4096
+supports_tensor = true
+
+[metadata.storage_size]
+in_bytes = 16882073600
--- a/resources/model_cards/llama-3.1-8b.toml
+++ b/resources/model_cards/llama-3.1-8b.toml
@@ -0,0 +1,15 @@
+short_id = "llama-3.1-8b"
+model_id = "mlx-community/Meta-Llama-3.1-8B-Instruct-4bit"
+name = "Llama 3.1 8B (4-bit)"
+description = "Llama 3.1 is a large language model trained on the Llama 3.1 dataset."
+tags = []
+
+[metadata]
+model_id = "mlx-community/Meta-Llama-3.1-8B-Instruct-4bit"
+pretty_name = "Llama 3.1 8B (4-bit)"
+n_layers = 32
+hidden_size = 4096
+supports_tensor = true
+
+[metadata.storage_size]
+in_bytes = 4637851648
--- a/resources/model_cards/llama-3.2-1b.toml
+++ b/resources/model_cards/llama-3.2-1b.toml
@@ -0,0 +1,15 @@
+short_id = "llama-3.2-1b"
+model_id = "mlx-community/Llama-3.2-1B-Instruct-4bit"
+name = "Llama 3.2 1B (4-bit)"
+description = "Llama 3.2 is a large language model trained on the Llama 3.2 dataset."
+tags = []
+
+[metadata]
+model_id = "mlx-community/Llama-3.2-1B-Instruct-4bit"
+pretty_name = "Llama 3.2 1B (4-bit)"
+n_layers = 16
+hidden_size = 2048
+supports_tensor = true
+
+[metadata.storage_size]
+in_bytes = 729808896
--- a/resources/model_cards/llama-3.2-3b-8bit.toml
+++ b/resources/model_cards/llama-3.2-3b-8bit.toml
@@ -0,0 +1,15 @@
+short_id = "llama-3.2-3b-8bit"
+model_id = "mlx-community/Llama-3.2-3B-Instruct-8bit"
+name = "Llama 3.2 3B (8-bit)"
+description = "Llama 3.2 is a large language model trained on the Llama 3.2 dataset."
+tags = []
+
+[metadata]
+model_id = "mlx-community/Llama-3.2-3B-Instruct-8bit"
+pretty_name = "Llama 3.2 3B (8-bit)"
+n_layers = 28
+hidden_size = 3072
+supports_tensor = true
+
+[metadata.storage_size]
+in_bytes = 3501195264
--- a/resources/model_cards/llama-3.2-3b.toml
+++ b/resources/model_cards/llama-3.2-3b.toml
@@ -0,0 +1,15 @@
+short_id = "llama-3.2-3b"
+model_id = "mlx-community/Llama-3.2-3B-Instruct-4bit"
+name = "Llama 3.2 3B (4-bit)"
+description = "Llama 3.2 is a large language model trained on the Llama 3.2 dataset."
+tags = []
+
+[metadata]
+model_id = "mlx-community/Llama-3.2-3B-Instruct-4bit"
+pretty_name = "Llama 3.2 3B (4-bit)"
+n_layers = 28
+hidden_size = 3072
+supports_tensor = true
+
+[metadata.storage_size]
+in_bytes = 1863319552
--- a/resources/model_cards/llama-3.3-70b-8bit.toml
+++ b/resources/model_cards/llama-3.3-70b-8bit.toml
@@ -0,0 +1,15 @@
+short_id = "llama-3.3-70b-8bit"
+model_id = "mlx-community/Llama-3.3-70B-Instruct-8bit"
+name = "Llama 3.3 70B (8-bit)"
+description = "The Meta Llama 3.3 multilingual large language model (LLM) is an instruction tuned generative model in 70B (text in/text out)"
+tags = []
+
+[metadata]
+model_id = "mlx-community/Llama-3.3-70B-Instruct-8bit"
+pretty_name = "Llama 3.3 70B (8-bit)"
+n_layers = 80
+hidden_size = 8192
+supports_tensor = true
+
+[metadata.storage_size]
+in_bytes = 76799803392
--- a/resources/model_cards/llama-3.3-70b-fp16.toml
+++ b/resources/model_cards/llama-3.3-70b-fp16.toml
@@ -0,0 +1,15 @@
+short_id = "llama-3.3-70b-fp16"
+model_id = "mlx-community/llama-3.3-70b-instruct-fp16"
+name = "Llama 3.3 70B (FP16)"
+description = "The Meta Llama 3.3 multilingual large language model (LLM) is an instruction tuned generative model in 70B (text in/text out)"
+tags = []
+
+[metadata]
+model_id = "mlx-community/llama-3.3-70b-instruct-fp16"
+pretty_name = "Llama 3.3 70B (FP16)"
+n_layers = 80
+hidden_size = 8192
+supports_tensor = true
+
+[metadata.storage_size]
+in_bytes = 144383672320
--- a/resources/model_cards/llama-3.3-70b.toml
+++ b/resources/model_cards/llama-3.3-70b.toml
@@ -0,0 +1,15 @@
+short_id = "llama-3.3-70b"
+model_id = "mlx-community/Llama-3.3-70B-Instruct-4bit"
+name = "Llama 3.3 70B (4-bit)"
+description = "The Meta Llama 3.3 multilingual large language model (LLM) is an instruction tuned generative model in 70B (text in/text out)"
+tags = []
+
+[metadata]
+model_id = "mlx-community/Llama-3.3-70B-Instruct-4bit"
+pretty_name = "Llama 3.3 70B"
+n_layers = 80
+hidden_size = 8192
+supports_tensor = true
+
+[metadata.storage_size]
+in_bytes = 40652242944
--- a/resources/model_cards/minimax-m2.1-3bit.toml
+++ b/resources/model_cards/minimax-m2.1-3bit.toml
@@ -0,0 +1,15 @@
+short_id = "minimax-m2.1-3bit"
+model_id = "mlx-community/MiniMax-M2.1-3bit"
+name = "MiniMax M2.1 3bit"
+description = "MiniMax M2.1 3bit"
+tags = []
+
+[metadata]
+model_id = "mlx-community/MiniMax-M2.1-3bit"
+pretty_name = "MiniMax M2.1 3bit"
+n_layers = 61
+hidden_size = 3072
+supports_tensor = true
+
+[metadata.storage_size]
+in_bytes = 100086644736
--- a/resources/model_cards/minimax-m2.1-8bit.toml
+++ b/resources/model_cards/minimax-m2.1-8bit.toml
@@ -0,0 +1,15 @@
+short_id = "minimax-m2.1-8bit"
+model_id = "mlx-community/MiniMax-M2.1-8bit"
+name = "MiniMax M2.1 8bit"
+description = "MiniMax M2.1 8bit"
+tags = []
+
+[metadata]
+model_id = "mlx-community/MiniMax-M2.1-8bit"
+pretty_name = "MiniMax M2.1 8bit"
+n_layers = 61
+hidden_size = 3072
+supports_tensor = true
+
+[metadata.storage_size]
+in_bytes = 242986745856
--- a/resources/model_cards/qwen3-0.6b-8bit.toml
+++ b/resources/model_cards/qwen3-0.6b-8bit.toml
@@ -0,0 +1,15 @@
+short_id = "qwen3-0.6b-8bit"
+model_id = "mlx-community/Qwen3-0.6B-8bit"
+name = "Qwen3 0.6B (8-bit)"
+description = "Qwen3 0.6B is a large language model trained on the Qwen3 0.6B dataset."
+tags = []
+
+[metadata]
+model_id = "mlx-community/Qwen3-0.6B-8bit"
+pretty_name = "Qwen3 0.6B (8-bit)"
+n_layers = 28
+hidden_size = 1024
+supports_tensor = false
+
+[metadata.storage_size]
+in_bytes = 698351616
--- a/resources/model_cards/qwen3-0.6b.toml
+++ b/resources/model_cards/qwen3-0.6b.toml
@@ -0,0 +1,15 @@
+short_id = "qwen3-0.6b"
+model_id = "mlx-community/Qwen3-0.6B-4bit"
+name = "Qwen3 0.6B (4-bit)"
+description = "Qwen3 0.6B is a large language model trained on the Qwen3 0.6B dataset."
+tags = []
+
+[metadata]
+model_id = "mlx-community/Qwen3-0.6B-4bit"
+pretty_name = "Qwen3 0.6B (4-bit)"
+n_layers = 28
+hidden_size = 1024
+supports_tensor = false
+
+[metadata.storage_size]
+in_bytes = 342884352
--- a/resources/model_cards/qwen3-235b-a22b-4bit.toml
+++ b/resources/model_cards/qwen3-235b-a22b-4bit.toml
@@ -0,0 +1,15 @@
+short_id = "qwen3-235b-a22b-4bit"
+model_id = "mlx-community/Qwen3-235B-A22B-Instruct-2507-4bit"
+name = "Qwen3 235B A22B (4-bit)"
+description = "Qwen3 235B (Active 22B) is a large language model trained on the Qwen3 235B dataset."
+tags = []
+
+[metadata]
+model_id = "mlx-community/Qwen3-235B-A22B-Instruct-2507-4bit"
+pretty_name = "Qwen3 235B A22B (4-bit)"
+n_layers = 94
+hidden_size = 4096
+supports_tensor = true
+
+[metadata.storage_size]
+in_bytes = 141733920768
--- a/resources/model_cards/qwen3-235b-a22b-8bit.toml
+++ b/resources/model_cards/qwen3-235b-a22b-8bit.toml
@@ -0,0 +1,15 @@
+short_id = "qwen3-235b-a22b-8bit"
+model_id = "mlx-community/Qwen3-235B-A22B-Instruct-2507-8bit"
+name = "Qwen3 235B A22B (8-bit)"
+description = "Qwen3 235B (Active 22B) is a large language model trained on the Qwen3 235B dataset."
+tags = []
+
+[metadata]
+model_id = "mlx-community/Qwen3-235B-A22B-Instruct-2507-8bit"
+pretty_name = "Qwen3 235B A22B (8-bit)"
+n_layers = 94
+hidden_size = 4096
+supports_tensor = true
+
+[metadata.storage_size]
+in_bytes = 268435456000
--- a/resources/model_cards/qwen3-30b-8bit.toml
+++ b/resources/model_cards/qwen3-30b-8bit.toml
@@ -0,0 +1,15 @@
+short_id = "qwen3-30b-8bit"
+model_id = "mlx-community/Qwen3-30B-A3B-8bit"
+name = "Qwen3 30B A3B (8-bit)"
+description = "Qwen3 30B is a large language model trained on the Qwen3 30B dataset."
+tags = []
+
+[metadata]
+model_id = "mlx-community/Qwen3-30B-A3B-8bit"
+pretty_name = "Qwen3 30B A3B (8-bit)"
+n_layers = 48
+hidden_size = 2048
+supports_tensor = true
+
+[metadata.storage_size]
+in_bytes = 33279705088
--- a/resources/model_cards/qwen3-30b.toml
+++ b/resources/model_cards/qwen3-30b.toml
@@ -0,0 +1,15 @@
+short_id = "qwen3-30b"
+model_id = "mlx-community/Qwen3-30B-A3B-4bit"
+name = "Qwen3 30B A3B (4-bit)"
+description = "Qwen3 30B is a large language model trained on the Qwen3 30B dataset."
+tags = []
+
+[metadata]
+model_id = "mlx-community/Qwen3-30B-A3B-4bit"
+pretty_name = "Qwen3 30B A3B (4-bit)"
+n_layers = 48
+hidden_size = 2048
+supports_tensor = true
+
+[metadata.storage_size]
+in_bytes = 17612931072
--- a/resources/model_cards/qwen3-80b-a3B-4bit.toml
+++ b/resources/model_cards/qwen3-80b-a3B-4bit.toml
@@ -0,0 +1,15 @@
+short_id = "qwen3-80b-a3B-4bit"
+model_id = "mlx-community/Qwen3-Next-80B-A3B-Instruct-4bit"
+name = "Qwen3 80B A3B (4-bit)"
+description = "Qwen3 80B"
+tags = []
+
+[metadata]
+model_id = "mlx-community/Qwen3-Next-80B-A3B-Instruct-4bit"
+pretty_name = "Qwen3 80B A3B (4-bit)"
+n_layers = 48
+hidden_size = 2048
+supports_tensor = true
+
+[metadata.storage_size]
+in_bytes = 46976204800
--- a/resources/model_cards/qwen3-80b-a3B-8bit.toml
+++ b/resources/model_cards/qwen3-80b-a3B-8bit.toml
@@ -0,0 +1,15 @@
+short_id = "qwen3-80b-a3B-8bit"
+model_id = "mlx-community/Qwen3-Next-80B-A3B-Instruct-8bit"
+name = "Qwen3 80B A3B (8-bit)"
+description = "Qwen3 80B"
+tags = []
+
+[metadata]
+model_id = "mlx-community/Qwen3-Next-80B-A3B-Instruct-8bit"
+pretty_name = "Qwen3 80B A3B (8-bit)"
+n_layers = 48
+hidden_size = 2048
+supports_tensor = true
+
+[metadata.storage_size]
+in_bytes = 88814387200
--- a/resources/model_cards/qwen3-80b-a3B-thinking-4bit.toml
+++ b/resources/model_cards/qwen3-80b-a3B-thinking-4bit.toml
@@ -0,0 +1,15 @@
+short_id = "qwen3-80b-a3B-thinking-4bit"
+model_id = "mlx-community/Qwen3-Next-80B-A3B-Thinking-4bit"
+name = "Qwen3 80B A3B Thinking (4-bit)"
+description = "Qwen3 80B Reasoning model"
+tags = []
+
+[metadata]
+model_id = "mlx-community/Qwen3-Next-80B-A3B-Thinking-4bit"
+pretty_name = "Qwen3 80B A3B (4-bit)"
+n_layers = 48
+hidden_size = 2048
+supports_tensor = true
+
+[metadata.storage_size]
+in_bytes = 88814387200
--- a/resources/model_cards/qwen3-80b-a3B-thinking-8bit.toml
+++ b/resources/model_cards/qwen3-80b-a3B-thinking-8bit.toml
@@ -0,0 +1,15 @@
+short_id = "qwen3-80b-a3B-thinking-8bit"
+model_id = "mlx-community/Qwen3-Next-80B-A3B-Thinking-8bit"
+name = "Qwen3 80B A3B Thinking (8-bit)"
+description = "Qwen3 80B Reasoning model"
+tags = []
+
+[metadata]
+model_id = "mlx-community/Qwen3-Next-80B-A3B-Thinking-8bit"
+pretty_name = "Qwen3 80B A3B (8-bit)"
+n_layers = 48
+hidden_size = 2048
+supports_tensor = true
+
+[metadata.storage_size]
+in_bytes = 88814387200
--- a/resources/model_cards/qwen3-coder-480b-a35b-4bit.toml
+++ b/resources/model_cards/qwen3-coder-480b-a35b-4bit.toml
@@ -0,0 +1,15 @@
+short_id = "qwen3-coder-480b-a35b-4bit"
+model_id = "mlx-community/Qwen3-Coder-480B-A35B-Instruct-4bit"
+name = "Qwen3 Coder 480B A35B (4-bit)"
+description = "Qwen3 Coder 480B (Active 35B) is a large language model trained on the Qwen3 Coder 480B dataset."
+tags = []
+
+[metadata]
+model_id = "mlx-community/Qwen3-Coder-480B-A35B-Instruct-4bit"
+pretty_name = "Qwen3 Coder 480B A35B (4-bit)"
+n_layers = 62
+hidden_size = 6144
+supports_tensor = true
+
+[metadata.storage_size]
+in_bytes = 289910292480
--- a/resources/model_cards/qwen3-coder-480b-a35b-8bit.toml
+++ b/resources/model_cards/qwen3-coder-480b-a35b-8bit.toml
@@ -0,0 +1,15 @@
+short_id = "qwen3-coder-480b-a35b-8bit"
+model_id = "mlx-community/Qwen3-Coder-480B-A35B-Instruct-8bit"
+name = "Qwen3 Coder 480B A35B (8-bit)"
+description = "Qwen3 Coder 480B (Active 35B) is a large language model trained on the Qwen3 Coder 480B dataset."
+tags = []
+
+[metadata]
+model_id = "mlx-community/Qwen3-Coder-480B-A35B-Instruct-8bit"
+pretty_name = "Qwen3 Coder 480B A35B (8-bit)"
+n_layers = 62
+hidden_size = 6144
+supports_tensor = true
+
+[metadata.storage_size]
+in_bytes = 579820584960
--- a/rust/parts.nix
+++ b/rust/parts.nix
@@ -0,0 +1,145 @@
+{ inputs, ... }:
+{
+  perSystem =
+    { config, self', inputs', pkgs, lib, ... }:
+    let
+      # Fenix nightly toolchain with all components
+      fenixPkgs = inputs'.fenix.packages;
+      rustToolchain = fenixPkgs.complete.withComponents [
+        "cargo"
+        "rustc"
+        "clippy"
+        "rustfmt"
+        "rust-src"
+        "rust-analyzer"
+      ];
+
+      # Crane with fenix toolchain
+      craneLib = (inputs.crane.mkLib pkgs).overrideToolchain rustToolchain;
+
+      # Source filtering - only include rust/ directory and root Cargo files
+      # This ensures changes to Python/docs/etc don't trigger Rust rebuilds
+      src = lib.cleanSourceWith {
+        src = inputs.self;
+        filter =
+          path: type:
+          let
+            baseName = builtins.baseNameOf path;
+            parentDir = builtins.dirOf path;
+            inRustDir =
+              (lib.hasInfix "/rust/" path)
+              || (lib.hasSuffix "/rust" parentDir)
+              || (baseName == "rust" && type == "directory");
+            isRootCargoFile =
+              (baseName == "Cargo.toml" || baseName == "Cargo.lock")
+              && (builtins.dirOf path == toString inputs.self);
+          in
+          isRootCargoFile
+          || (inRustDir && (craneLib.filterCargoSources path type || lib.hasSuffix ".toml" path || lib.hasSuffix ".md" path));
+      };
+
+      # Common arguments for all Rust builds
+      commonArgs = {
+        inherit src;
+        pname = "exo-rust";
+        version = "0.0.1";
+        strictDeps = true;
+
+        nativeBuildInputs = [
+          pkgs.pkg-config
+          pkgs.python313 # Required for pyo3-build-config
+        ];
+
+        buildInputs = [
+          pkgs.openssl
+          pkgs.python313 # Required for pyo3 tests
+        ];
+
+        OPENSSL_NO_VENDOR = "1";
+
+        # Required for pyo3 tests to find libpython
+        LD_LIBRARY_PATH = lib.makeLibraryPath [ pkgs.python313 ];
+      };
+
+      # Build dependencies once for caching
+      cargoArtifacts = craneLib.buildDepsOnly (
+        commonArgs
+        // {
+          cargoExtraArgs = "--workspace";
+        }
+      );
+    in
+    {
+      # Export toolchain for use in treefmt and devShell
+      options.rust = {
+        toolchain = lib.mkOption {
+          type = lib.types.package;
+          default = rustToolchain;
+          description = "The Rust toolchain to use";
+        };
+      };
+
+      config = {
+        packages = {
+          # Python bindings wheel via maturin
+          exo_pyo3_bindings = craneLib.buildPackage (
+            commonArgs
+            // {
+              inherit cargoArtifacts;
+              pname = "exo_pyo3_bindings";
+
+              nativeBuildInputs = commonArgs.nativeBuildInputs ++ [
+                pkgs.maturin
+              ];
+
+              buildPhaseCargoCommand = ''
+                maturin build \
+                  --release \
+                  --manylinux off \
+                  --manifest-path rust/exo_pyo3_bindings/Cargo.toml \
+                  --features "pyo3/extension-module,pyo3/experimental-async" \
+                  --interpreter ${pkgs.python313}/bin/python \
+                  --out dist
+              '';
+
+              # Don't use crane's default install behavior
+              doNotPostBuildInstallCargoBinaries = true;
+
+              installPhaseCommand = ''
+                mkdir -p $out
+                cp dist/*.whl $out/
+              '';
+            }
+          );
+        };
+
+        checks = {
+          # Full workspace build (all crates)
+          cargo-build = craneLib.buildPackage (
+            commonArgs
+            // {
+              inherit cargoArtifacts;
+              cargoExtraArgs = "--workspace";
+            }
+          );
+          # Run tests with nextest
+          cargo-nextest = craneLib.cargoNextest (
+            commonArgs
+            // {
+              inherit cargoArtifacts;
+              cargoExtraArgs = "--workspace";
+            }
+          );
+
+          # Build documentation
+          cargo-doc = craneLib.cargoDoc (
+            commonArgs
+            // {
+              inherit cargoArtifacts;
+              cargoExtraArgs = "--workspace";
+            }
+          );
+        };
+      };
+    };
+}
--- a/rust/system_custodian/Cargo.toml
+++ b/rust/system_custodian/Cargo.toml
@@ -1,47 +0,0 @@
-[package]
-name = "system_custodian"
-version = { workspace = true }
-edition = { workspace = true }
-publish = false
-
-[lib]
-doctest = false
-name = "system_custodian"
-path = "src/lib.rs"
-
-[[bin]]
-path = "src/bin/main.rs"
-name = "system_custodian"
-doc = false
-
-[lints]
-workspace = true
-
-[dependencies]
-# datastructures
-either = { workspace = true }
-
-# macro dependencies
-extend = { workspace = true }
-delegate = { workspace = true }
-impl-trait-for-tuples = { workspace = true }
-derive_more = { workspace = true }
-
-# async
-tokio = { workspace = true, features = ["full"] }
-futures = { workspace = true }
-futures-timer = { workspace = true }
-
-# utility dependencies
-util = { workspace = true }
-thiserror = { workspace = true }
-#internment = { workspace = true }
-#recursion = { workspace = true }
-#generativity = { workspace = true }
-#itertools = { workspace = true }
-tracing-subscriber = { version = "0.3.19", features = ["default", "env-filter"] }
-keccak-const = { workspace = true }
-
-# tracing/logging
-log = { workspace = true }
-
--- a/rust/system_custodian/src/bin/main.rs
+++ b/rust/system_custodian/src/bin/main.rs
@@ -1,4 +0,0 @@
-//! TODO: documentation
-//!
-
-fn main() {}
--- a/rust/system_custodian/src/lib.rs
+++ b/rust/system_custodian/src/lib.rs
@@ -1,69 +0,0 @@
-//! This crate defines the logic of, and ways to interact with, Exo's **_System Custodian_** daemon.
-//!
-//! The **_System Custodian_** daemon is supposed to be a long-living process that precedes the
-//! launch of the Exo application, and responsible for ensuring the system (configuration, settings,
-//! etc.) is in an appropriate state to facilitate the running of Exo application.
-//! The **_System Custodian_** daemon shall expose a [D-Bus](https://www.freedesktop.org/wiki/Software/dbus/)
-//! service which Exo application use to _control & query_ it.
-//!
-//! # Lifecycle
-//! When the Exo application starts, it will _wake_ the **_System Custodian_** daemon for the
-//! duration of its lifetime, and after it has terminated the daemon will go back to sleep. When
-//! the daemon wakes up, it will configure the system into a state suitable for the Exo Application;
-//! When the daemon goes to sleep, it will revert those changes as much as it can in case they were
-//! destructive to the user's pre-existing configurations.
-//!
-//! # Responsibilities
-//! TODO: these are purely on MacOS, but change to be more broad
-//! The **_System Custodian_** daemon is responsible for using System Configuration framework to
-//!  1. duplicate the current network set
-//!  2. modify existing services to turn on IPv6 if not there
-//!  3. remove any bridge services & add any missing services that AREN'T bridge
-//! TODO: In the future:
-//!  1. run a dummy AWDL service to [allow for macOS peer-to-peer wireless networking](https://yggdrasil-network.github.io/2019/08/19/awdl.html)
-//!  2. toggle some GPU/memory configurations to speed up GPU (ask Alex what those configurations are)
-//!  3. if we ever decide to provide our **own network interfaces** that abstract over some userland
-//!     logic, this would be the place to spin that up.
-//!
-//! Then it will watch the SCDynamicStore for:
-//!  1. all __actual__ network interfaces -> collect information on them e.g. their BSD name, MAC
-//!     address, MTU, IPv6 addresses, etc. -> and set up watchers/notifiers to inform the DBus
-//!     interface of any changes
-//!  2. watch for any __undesirable__ changes to configuration and revert it
-//!
-//! It should somehow (probably through system sockets and/or BSD interface) trigger IPv6 NDP on
-//! each of the interfaces & also listen to/query for any changes on the OS routing cache??
-//! Basically emulate the `ping6 ff02::1%enX` and `ndp -an` commands BUT BETTER!!!
-//!  1. all that info should coalesce back to the overall state colleted -> should be queryable
-//!     over D-Bus
-//! TODO:
-//!  1. we might potentially add to this step a handshake of some kind...? To ensure that we can
-//!     ACTUALLY communicate with that machine over that link over e.g. TCP, UDP, etc. Will the
-//!     handshake require to know Node ID? Will the handshake require heartbeats? Who knows...
-//!  2. if we ever decide to write proprietary L2/L3 protocols for quicker communication,
-//!     e.g. [AF_NDRV](https://www.zerotier.com/blog/how-zerotier-eliminated-kernel-extensions-on-macos/)
-//!     for raw ethernet frame communication, or even a [custom thunderbolt PCIe driver](https://developer.apple.com/documentation/pcidriverkit/creating-custom-pcie-drivers-for-thunderbolt-devices),
-//!     then this would be the place to carry out discovery and propper handshakes with devices
-//!     on the other end of the link.
-//!
-
-// enable Rust-unstable features for convenience
-#![feature(trait_alias)]
-#![feature(stmt_expr_attributes)]
-#![feature(type_alias_impl_trait)]
-#![feature(specialization)]
-#![feature(unboxed_closures)]
-#![feature(const_trait_impl)]
-#![feature(fn_traits)]
-
-pub(crate) mod private {
-    // sealed traits support
-    pub trait Sealed {}
-    impl<T: ?Sized> Sealed for T {}
-}
-
-/// Namespace for all the type/trait aliases used by this crate.
-pub(crate) mod alias {}
-
-/// Namespace for crate-wide extension traits/methods
-pub(crate) mod ext {}
--- a/src/exo/shared/models/model_cards.py
+++ b/src/exo/shared/models/model_cards.py
@@ -1,3 +1,6 @@
+from anyio import Path, open_file
+import tomlkit 
+
 from exo.shared.types.memory import Memory
 from exo.shared.types.models import ModelId, ModelMetadata
 from exo.utils.pydantic_ext import CamelCaseModel
@@ -11,35 +14,24 @@ class ModelCard(CamelCaseModel):
    tags: list[str]
    metadata: ModelMetadata

+    @staticmethod
+    async def load(path: Path) -> "ModelCard":
+        async with await open_file(path) as f:
+            data = await f.read()
+            py = tomlkit.loads(data)
+            return ModelCard.model_validate(py)
+
+    async def save(self, path: Path):
+        async with await open_file(path, "w") as f:
+            py = self.model_dump()
+            data = tomlkit.dumps(py) # pyright: ignore[reportUnknownMemberType]
+            await f.write(data)
+
+
+

 MODEL_CARDS: dict[str, ModelCard] = {
    # deepseek v3
-    # "deepseek-v3-0324:4bit": ModelCard(
-    #     short_id="deepseek-v3-0324:4bit",
-    #     model_id="mlx-community/DeepSeek-V3-0324-4bit",
-    #     name="DeepSeek V3 0324 (4-bit)",
-    #     description="""DeepSeek V3 is a large language model trained on the DeepSeek V3 dataset.""",
-    #     tags=[],
-    #     metadata=ModelMetadata(
-    #         model_id=ModelId("mlx-community/DeepSeek-V3-0324-4bit"),
-    #         pretty_name="DeepSeek V3 0324 (4-bit)",
-    #         storage_size=Memory.from_kb(409706307),
-    #         n_layers=61,
-    #     ),
-    # ),
-    # "deepseek-v3-0324": ModelCard(
-    #     short_id="deepseek-v3-0324",
-    #     model_id="mlx-community/DeepSeek-v3-0324-8bit",
-    #     name="DeepSeek V3 0324 (8-bit)",
-    #     description="""DeepSeek V3 is a large language model trained on the DeepSeek V3 dataset.""",
-    #     tags=[],
-    #     metadata=ModelMetadata(
-    #         model_id=ModelId("mlx-community/DeepSeek-v3-0324-8bit"),
-    #         pretty_name="DeepSeek V3 0324 (8-bit)",
-    #         storage_size=Memory.from_kb(754706307),
-    #         n_layers=61,
-    #     ),
-    # ),
    "deepseek-v3.1-4bit": ModelCard(
        short_id="deepseek-v3.1-4bit",
        model_id=ModelId("mlx-community/DeepSeek-V3.1-4bit"),
@@ -70,63 +62,6 @@ MODEL_CARDS: dict[str, ModelCard] = {
            supports_tensor=True,
        ),
    ),
-    # "deepseek-v3.2": ModelCard(
-    #     short_id="deepseek-v3.2",
-    #     model_id=ModelId("mlx-community/DeepSeek-V3.2-8bit"),
-    #     name="DeepSeek V3.2 (8-bit)",
-    #     description="""DeepSeek V3.2 is a large language model trained on the DeepSeek V3.2 dataset.""",
-    #     tags=[],
-    #     metadata=ModelMetadata(
-    #         model_id=ModelId("mlx-community/DeepSeek-V3.2-8bit"),
-    #         pretty_name="DeepSeek V3.2 (8-bit)",
-    #         storage_size=Memory.from_kb(754706307),
-    #         n_layers=61,
-    #         hidden_size=7168,
-    #     ),
-    # ),
-    # "deepseek-v3.2-4bit": ModelCard(
-    #     short_id="deepseek-v3.2-4bit",
-    #     model_id=ModelId("mlx-community/DeepSeek-V3.2-4bit"),
-    #     name="DeepSeek V3.2 (4-bit)",
-    #     description="""DeepSeek V3.2 is a large language model trained on the DeepSeek V3.2 dataset.""",
-    #     tags=[],
-    #     metadata=ModelMetadata(
-    #         model_id=ModelId("mlx-community/DeepSeek-V3.2-4bit"),
-    #         pretty_name="DeepSeek V3.2 (4-bit)",
-    #         storage_size=Memory.from_kb(754706307 // 2),  # TODO !!!!!
-    #         n_layers=61,
-    #         hidden_size=7168,
-    #     ),
-    # ),
-    # deepseek r1
-    # "deepseek-r1-0528-4bit": ModelCard(
-    #     short_id="deepseek-r1-0528-4bit",
-    #     model_id="mlx-community/DeepSeek-R1-0528-4bit",
-    #     name="DeepSeek-R1-0528 (4-bit)",
-    #     description="""DeepSeek R1 is a large language model trained on the DeepSeek R1 dataset.""",
-    #     tags=[],
-    #     metadata=ModelMetadata(
-    #         model_id=ModelId("mlx-community/DeepSeek-R1-0528-4bit"),
-    #         pretty_name="DeepSeek R1 671B (4-bit)",
-    #         storage_size=Memory.from_kb(409706307),
-    #         n_layers=61,
-    #         hidden_size=7168,
-    #     ),
-    # ),
-    # "deepseek-r1-0528": ModelCard(
-    #     short_id="deepseek-r1-0528",
-    #     model_id="mlx-community/DeepSeek-R1-0528-8bit",
-    #     name="DeepSeek-R1-0528 (8-bit)",
-    #     description="""DeepSeek R1 is a large language model trained on the DeepSeek R1 dataset.""",
-    #     tags=[],
-    #     metadata=ModelMetadata(
-    #         model_id=ModelId("mlx-community/DeepSeek-R1-0528-8bit"),
-    #         pretty_name="DeepSeek R1 671B (8-bit)",
-    #         storage_size=Memory.from_bytes(754998771712),
-    #         n_layers=61,
-    # .       hidden_size=7168,
-    #     ),
-    # ),
    # kimi k2
    "kimi-k2-instruct-4bit": ModelCard(
        short_id="kimi-k2-instruct-4bit",
@@ -523,8 +458,9 @@ MODEL_CARDS: dict[str, ModelCard] = {
            supports_tensor=True,
        ),
    ),
-    # Needs to be quantized g32 or g16.
+    # glm 4.5
    "glm-4.5-air-8bit": ModelCard(
+        # Needs to be quantized g32 or g16 to work with tensor parallel
        short_id="glm-4.5-air-8bit",
        model_id=ModelId("mlx-community/GLM-4.5-Air-8bit"),
        name="GLM 4.5 Air 8bit",
@@ -554,19 +490,81 @@ MODEL_CARDS: dict[str, ModelCard] = {
            supports_tensor=True,
        ),
    ),
-    # "devstral-2-123b-instruct-2512-8bit": ModelCard(
-    #     short_id="devstral-2-123b-instruct-2512-8bit",
-    #     model_id=ModelId("mlx-community/Devstral-2-123B-Instruct-2512-8bit"),
-    #     name="Devstral 2 123B Instruct 2512 (8-bit, MLX)",
-    #     description="""Mistral AI's Devstral 2 123B Instruct (2512) is an agentic coding model.""",
-    #     tags=[],
-    #     metadata=ModelMetadata(
-    #         model_id=ModelId("mlx-community/Devstral-2-123B-Instruct-2512-8bit"),
-    #         pretty_name="Devstral 2 123B Instruct 2512 (8-bit, MLX)",
-    #         storage_size=Memory.from_kb(133_000_000),
-    #         n_layers=88,
-    #         hidden_size=12288,
-    #         supports_tensor=True,
-    #     ),
-    # ),
+    # glm 4.7
+    "glm-4.7-4bit": ModelCard(
+        short_id="glm-4.7-4bit",
+        model_id=ModelId("mlx-community/GLM-4.7-4bit"),
+        name="GLM 4.7 4bit",
+        description="GLM 4.7 4bit",
+        tags=[],
+        metadata=ModelMetadata(
+            model_id=ModelId("mlx-community/GLM-4.7-4bit"),
+            pretty_name="GLM 4.7 4bit",
+            storage_size=Memory.from_bytes(198556925568),
+            n_layers=91,
+            hidden_size=5120,
+            supports_tensor=True,
+        ),
+    ),
+    "glm-4.7-6bit": ModelCard(
+        short_id="glm-4.7-6bit",
+        model_id=ModelId("mlx-community/GLM-4.7-6bit"),
+        name="GLM 4.7 6bit",
+        description="GLM 4.7 6bit",
+        tags=[],
+        metadata=ModelMetadata(
+            model_id=ModelId("mlx-community/GLM-4.7-6bit"),
+            pretty_name="GLM 4.7 6bit",
+            storage_size=Memory.from_bytes(286737579648),
+            n_layers=91,
+            hidden_size=5120,
+            supports_tensor=True,
+        ),
+    ),
+    "glm-4.7-8bit-gs32": ModelCard(
+        short_id="glm-4.7-8bit-gs32",
+        model_id=ModelId("mlx-community/GLM-4.7-8bit-gs32"),
+        name="GLM 4.7 8bit (gs32)",
+        description="GLM 4.7 8bit (gs32)",
+        tags=[],
+        metadata=ModelMetadata(
+            model_id=ModelId("mlx-community/GLM-4.7-8bit-gs32"),
+            pretty_name="GLM 4.7 8bit (gs32)",
+            storage_size=Memory.from_bytes(396963397248),
+            n_layers=91,
+            hidden_size=5120,
+            supports_tensor=True,
+        ),
+    ),
+    # minimax-m2
+    "minimax-m2.1-8bit": ModelCard(
+        short_id="minimax-m2.1-8bit",
+        model_id=ModelId("mlx-community/MiniMax-M2.1-8bit"),
+        name="MiniMax M2.1 8bit",
+        description="MiniMax M2.1 8bit",
+        tags=[],
+        metadata=ModelMetadata(
+            model_id=ModelId("mlx-community/MiniMax-M2.1-8bit"),
+            pretty_name="MiniMax M2.1 8bit",
+            storage_size=Memory.from_bytes(242986745856),
+            n_layers=61,
+            hidden_size=3072,
+            supports_tensor=True,
+        ),
+    ),
+    "minimax-m2.1-3bit": ModelCard(
+        short_id="minimax-m2.1-3bit",
+        model_id=ModelId("mlx-community/MiniMax-M2.1-3bit"),
+        name="MiniMax M2.1 3bit",
+        description="MiniMax M2.1 3bit",
+        tags=[],
+        metadata=ModelMetadata(
+            model_id=ModelId("mlx-community/MiniMax-M2.1-3bit"),
+            pretty_name="MiniMax M2.1 3bit",
+            storage_size=Memory.from_bytes(100086644736),
+            n_layers=61,
+            hidden_size=3072,
+            supports_tensor=True,
+        ),
+    ),
 }
--- a/src/exo/worker/engines/mlx/auto_parallel.py
+++ b/src/exo/worker/engines/mlx/auto_parallel.py
@@ -10,18 +10,24 @@ from mlx.nn.layers.distributed import (
    shard_linear,
    sum_gradients,
 )
-from mlx_lm.models.cache import (
-    _BaseCache,  # pyright: ignore[reportPrivateUsage]
-)
 from mlx_lm.models.deepseek_v3 import DeepseekV3MLP
 from mlx_lm.models.deepseek_v3 import Model as DeepseekV3Model
+from mlx_lm.models.deepseek_v32 import DeepseekV32MLP
+from mlx_lm.models.deepseek_v32 import Model as DeepseekV32Model
+from mlx_lm.models.glm4_moe import Model as Glm4MoeModel
+from mlx_lm.models.glm4_moe import MoE
+from mlx_lm.models.gpt_oss import GptOssMoeModel
+from mlx_lm.models.gpt_oss import Model as GptOssModel
 from mlx_lm.models.llama import Model as LlamaModel
+from mlx_lm.models.minimax import Model as MiniMaxModel
+from mlx_lm.models.ministral3 import Model as Ministral3Model
 from mlx_lm.models.qwen3_moe import Model as Qwen3MoeModel
 from mlx_lm.models.qwen3_moe import Qwen3MoeSparseMoeBlock
+from mlx_lm.models.qwen3_next import Model as Qwen3NextModel
+from mlx_lm.models.qwen3_next import Qwen3NextSparseMoeBlock

-from exo.shared.types.worker.shards import (
-    PipelineShardMetadata,
-)
+from exo.shared.logging import logger
+from exo.shared.types.worker.shards import PipelineShardMetadata


 class _LayerCallable(Protocol):
@@ -91,8 +97,6 @@ class PipelineLastLayer(CustomMlxLayer):
            x, *args, **kwargs
        ).arguments.get("cache", None)

-        assert cache is None or issubclass(type(cache), _BaseCache)  # type: ignore
-
        output: mx.array = self.original_layer(x, *args, **kwargs)

        if self.r != self.s - 1:
@@ -100,7 +104,6 @@ class PipelineLastLayer(CustomMlxLayer):
                output, (self.r + 1) % self.s, group=self.group
            )
            if cache is not None:
-                # This change happened upstream - check out mlx github somewhere??
                cache.keys = mx.depends(cache.keys, output)  # type: ignore[reportUnknownMemberType]

        output = mx.distributed.all_gather(output, group=self.group)[-output.shape[0] :]
@@ -132,24 +135,6 @@ def _get_layers(inner_model_instance: nn.Module) -> list[_LayerCallable]:
    return layers


-def _set_layers(model: nn.Module, layers: list[_LayerCallable]) -> None:
-    inner_model_instance = _inner_model(model)
-    if hasattr(inner_model_instance, "layers"):
-        inner_model_instance.layers = layers
-
-        # Update DeepSeek V3 specific parameters when layers are shrunk
-        if isinstance(model, DeepseekV3Model) and hasattr(
-            inner_model_instance, "num_layers"
-        ):
-            inner_model_instance.start_idx = 0
-            inner_model_instance.end_idx = len(layers)
-            inner_model_instance.num_layers = len(layers)
-    elif hasattr(inner_model_instance, "h"):
-        inner_model_instance.h = layers
-    else:
-        raise ValueError("Model must have either a 'layers' or 'h' attribute")
-
-
 def pipeline_auto_parallel(
    model: nn.Module,
    group: mx.distributed.Group,
@@ -165,8 +150,7 @@ def pipeline_auto_parallel(
    """
    inner_model_instance: nn.Module = _inner_model(model)

-    # Handle both model.layers and model.h cases
-    layers: list[_LayerCallable] = _get_layers(inner_model_instance)
+    layers = _get_layers(inner_model_instance)

    start_layer, end_layer = model_shard_meta.start_layer, model_shard_meta.end_layer
    device_rank, world_size = model_shard_meta.device_rank, model_shard_meta.world_size
@@ -180,6 +164,17 @@ def pipeline_auto_parallel(
        group=group,
    )

+    if isinstance(inner_model_instance, GptOssMoeModel):
+        inner_model_instance.layer_types = inner_model_instance.layer_types[  # type: ignore
+            start_layer:end_layer
+        ]
+        inner_model_instance.swa_idx = inner_model_instance.layer_types.index(  # type: ignore
+            "sliding_attention"
+        )
+        inner_model_instance.ga_idx = inner_model_instance.layer_types.index(  # type: ignore
+            "full_attention"
+        )
+
    _set_layers(model, layers)

    assert isinstance(layers, list), (
@@ -204,18 +199,44 @@ def tensor_auto_parallel(
        group=group,
    )

+    segments: int = 1
+
+    def _all_to_sharded(path: str, weight: mx.array):
+        if path.endswith("bias"):
+            logger.info(f"Sharding bias for {path} - all to sharded")
+            return weight.ndim - 1, segments
+        return max(weight.ndim - 2, 0), segments
+
    all_to_sharded_linear_in_place = partial(
        shard_inplace,
-        sharding="all-to-sharded",
-        group=group,
-    )
-    sharded_to_all_linear_in_place = partial(
-        shard_inplace,
-        sharding="sharded-to-all",
+        sharding=_all_to_sharded,  # type: ignore
        group=group,
    )

-    if isinstance(model, LlamaModel):
+    n = group.size()
+
+    def _sharded_to_all(path: str, weight: mx.array):
+        if path.endswith("bias"):
+            logger.info(f"Sharding bias for {path} - sharded to all")
+            weight /= n
+            return None
+        return -1, segments
+
+    sharded_to_all_linear_in_place = partial(
+        shard_inplace,
+        sharding=_sharded_to_all,  # type: ignore
+        group=group,
+    )
+
+    if hasattr(model, "shard"):
+        try:
+            model.shard(group)  # type: ignore
+            return model
+        except (AttributeError, TypeError, NameError):
+            pass
+
+    if isinstance(model, (LlamaModel, Ministral3Model)):
+        logger.warning("shouldn't be hit - upstream sharding exists")
        tensor_parallel_sharding_strategy = LlamaShardingStrategy(
            group,
            all_to_sharded_linear,
@@ -223,7 +244,8 @@ def tensor_auto_parallel(
            all_to_sharded_linear_in_place,
            sharded_to_all_linear_in_place,
        )
-    elif isinstance(model, DeepseekV3Model):
+    elif isinstance(model, (DeepseekV3Model, DeepseekV32Model)):
+        logger.warning("shouldn't be hit - upstream sharding exists")
        tensor_parallel_sharding_strategy = DeepSeekShardingStrategy(
            group,
            all_to_sharded_linear,
@@ -231,7 +253,15 @@ def tensor_auto_parallel(
            all_to_sharded_linear_in_place,
            sharded_to_all_linear_in_place,
        )
-    elif isinstance(model, Qwen3MoeModel):
+    elif isinstance(model, MiniMaxModel):
+        tensor_parallel_sharding_strategy = MiniMaxShardingStrategy(
+            group,
+            all_to_sharded_linear,
+            sharded_to_all_linear,
+            all_to_sharded_linear_in_place,
+            sharded_to_all_linear_in_place,
+        )
+    elif isinstance(model, (Qwen3MoeModel, Glm4MoeModel, Qwen3NextModel)):
        tensor_parallel_sharding_strategy = QwenShardingStrategy(
            group,
            all_to_sharded_linear,
@@ -239,6 +269,15 @@ def tensor_auto_parallel(
            all_to_sharded_linear_in_place,
            sharded_to_all_linear_in_place,
        )
+    elif isinstance(model, GptOssModel):
+        tensor_parallel_sharding_strategy = GptOssShardingStrategy(
+            group,
+            all_to_sharded_linear,
+            sharded_to_all_linear,
+            all_to_sharded_linear_in_place,
+            sharded_to_all_linear_in_place,
+        )
+
    else:
        raise ValueError(f"Unsupported model type: {type(model)}")

@@ -284,6 +323,32 @@ class LlamaShardingStrategy(TensorParallelShardingStrategy):
        return model


+def _set_layers(model: nn.Module, layers: list[_LayerCallable]) -> None:
+    inner_model_instance = _inner_model(model)
+    if hasattr(inner_model_instance, "layers"):
+        inner_model_instance.layers = layers
+
+        # Update DeepSeek V3 specific parameters when layers are shrunk
+        if isinstance(
+            model, (DeepseekV3Model, DeepseekV32Model, Glm4MoeModel)
+        ) and hasattr(inner_model_instance, "num_layers"):
+            logger.info(
+                f"Setting num_layers to {len(layers)} for model {model.model.__class__.__name__}"
+            )
+            inner_model_instance.start_idx = 0
+            inner_model_instance.end_idx = len(layers)
+            inner_model_instance.num_layers = len(layers)
+        elif isinstance(model, Qwen3MoeModel):
+            logger.info(
+                f"Setting num_hidden_layers to {len(layers)} for model {model.model.__class__.__name__}"
+            )
+            inner_model_instance.num_hidden_layers = len(layers)
+    elif hasattr(inner_model_instance, "h"):
+        inner_model_instance.h = layers
+    else:
+        raise ValueError("Model must have either a 'layers' or 'h' attribute")
+
+
 class DeepSeekShardingStrategy(TensorParallelShardingStrategy):
    def shard_model(self, model: nn.Module) -> nn.Module:
        model = cast(DeepseekV3Model, model)
@@ -304,7 +369,7 @@ class DeepSeekShardingStrategy(TensorParallelShardingStrategy):
            layer.self_attn.num_heads //= self.N

            # Shard the MLP
-            if isinstance(layer.mlp, DeepseekV3MLP):
+            if isinstance(layer.mlp, (DeepseekV3MLP, DeepseekV32MLP)):
                layer.mlp.gate_proj = self.all_to_sharded_linear(layer.mlp.gate_proj)
                layer.mlp.down_proj = self.sharded_to_all_linear(layer.mlp.down_proj)
                layer.mlp.up_proj = self.all_to_sharded_linear(layer.mlp.up_proj)
@@ -338,6 +403,35 @@ class ShardedDeepseekV3MoE(CustomMlxLayer):
        return y


+class MiniMaxShardingStrategy(TensorParallelShardingStrategy):
+    def shard_model(self, model: nn.Module) -> nn.Module:
+        model = cast(MiniMaxModel, model)
+        for layer in model.layers:
+            # Shard the self attention
+            layer.self_attn.q_proj = self.all_to_sharded_linear(layer.self_attn.q_proj)
+            layer.self_attn.k_proj = self.all_to_sharded_linear(layer.self_attn.k_proj)
+            layer.self_attn.v_proj = self.all_to_sharded_linear(layer.self_attn.v_proj)
+            layer.self_attn.o_proj = self.sharded_to_all_linear(layer.self_attn.o_proj)
+            layer.self_attn.num_attention_heads //= self.N
+            layer.self_attn.num_key_value_heads //= self.N
+
+            # Shard the MoE. Shard in place since the MoE should be responsible
+            # for aggregating the results.
+            self.all_to_sharded_linear_in_place(
+                layer.block_sparse_moe.switch_mlp.gate_proj
+            )
+            self.sharded_to_all_linear_in_place(
+                layer.block_sparse_moe.switch_mlp.down_proj
+            )
+            self.all_to_sharded_linear_in_place(
+                layer.block_sparse_moe.switch_mlp.up_proj
+            )
+            layer.block_sparse_moe = ShardedQwenMoE(layer.block_sparse_moe)  # pyright: ignore[reportAttributeAccessIssue, reportArgumentType]
+            layer.block_sparse_moe.sharding_group = self.group
+
+        return model
+
+
 class QwenShardingStrategy(TensorParallelShardingStrategy):
    def shard_model(self, model: nn.Module) -> nn.Module:
        model = cast(Qwen3MoeModel, model)
@@ -352,11 +446,13 @@ class QwenShardingStrategy(TensorParallelShardingStrategy):

            # Shard the MoE. Shard in place since the MoE should be responsible
            # for aggregating the results.
-            if isinstance(layer.mlp, Qwen3MoeSparseMoeBlock):
+            if isinstance(
+                layer.mlp, (Qwen3MoeSparseMoeBlock, MoE, Qwen3NextSparseMoeBlock)
+            ):
                self.all_to_sharded_linear_in_place(layer.mlp.switch_mlp.gate_proj)
                self.sharded_to_all_linear_in_place(layer.mlp.switch_mlp.down_proj)
                self.all_to_sharded_linear_in_place(layer.mlp.switch_mlp.up_proj)
-                layer.mlp = ShardedQwenMoE(layer.mlp)  # type: ignore
+                layer.mlp = ShardedQwenMoE(layer.mlp)  # pyright: ignore[reportAttributeAccessIssue, reportArgumentType]
                layer.mlp.sharding_group = self.group

            # Shard the MLP
@@ -380,3 +476,50 @@ class ShardedQwenMoE(CustomMlxLayer):
        if self.sharding_group is not None:
            y = mx.distributed.all_sum(y, group=self.sharding_group)
        return y
+
+
+class GptOssShardingStrategy(TensorParallelShardingStrategy):
+    def shard_model(self, model: nn.Module) -> nn.Module:
+        model = cast(GptOssMoeModel, model)
+
+        for layer in model.layers:
+            layer.self_attn.q_proj = self.all_to_sharded_linear(layer.self_attn.q_proj)
+            layer.self_attn.k_proj = self.all_to_sharded_linear(layer.self_attn.k_proj)
+            layer.self_attn.v_proj = self.all_to_sharded_linear(layer.self_attn.v_proj)
+            layer.self_attn.o_proj = self.sharded_to_all_linear(layer.self_attn.o_proj)
+
+            layer.self_attn.num_attention_heads //= self.N
+            layer.self_attn.num_key_value_heads //= self.N
+            layer.self_attn.num_key_value_groups = (
+                layer.self_attn.num_attention_heads
+                // layer.self_attn.num_key_value_heads
+            )
+
+            layer.self_attn.sinks = layer.self_attn.sinks[
+                layer.self_attn.num_attention_heads
+                * self.group.rank() : layer.self_attn.num_attention_heads
+                * (self.group.rank() + 1)
+            ]
+
+            self.all_to_sharded_linear_in_place(layer.mlp.experts.gate_proj)
+            self.sharded_to_all_linear_in_place(layer.mlp.experts.down_proj)
+            self.all_to_sharded_linear_in_place(layer.mlp.experts.up_proj)
+
+            layer.mlp = ShardedGptOssMoE(layer.mlp)  # type: ignore
+            layer.mlp.sharding_group = self.group
+
+        return model
+
+
+class ShardedGptOssMoE(CustomMlxLayer):
+    def __init__(self, layer: nn.Module):
+        super().__init__(layer)
+        self.sharding_group: mx.distributed.Group | None = None
+
+    def __call__(self, x: mx.array) -> mx.array:
+        if self.sharding_group is not None:
+            x = sum_gradients(self.sharding_group)(x)
+        y = self.original_layer(x)
+        if self.sharding_group is not None:
+            y = mx.distributed.all_sum(y, group=self.sharding_group)
+        return y
--- a/tests/headless_runner.py
+++ b/tests/headless_runner.py
@@ -89,6 +89,12 @@ async def assert_downloads():
    await sd.ensure_shard(
        await build_full_shard(MODEL_CARDS["gpt-oss-20b-4bit"].model_id)
    )
+    await sd.ensure_shard(
+        await build_full_shard(MODEL_CARDS["glm-4.7-8bit-gs32"].model_id)
+    )
+    await sd.ensure_shard(
+        await build_full_shard(MODEL_CARDS["minimax-m2.1-8bit"].model_id)
+    )


 async def ring_backend(test: Tests):
--- a/uv.lock
+++ b/uv.lock
Author	SHA1	Message	Date
Evan	642b1bb1b4	migrate model cards to .toml files	2026-01-15 17:07:48 +00:00
Evan Quiney	c22dad8a7d	dashboard: add peer: true to package lock (#1162 ) this happens every time i run npm install - lets upstream it ## testing dashboard builds and renders	2026-01-15 17:01:43 +00:00
Evan	4bc4d50685	rust: remove dead code the system custodian has been made unnecessary with the swift app - we can remove it ## testing everything still builds	2026-01-15 16:51:46 +00:00
Jake Hillion	e0aab46fd8	model_cards.py: clean up commented out code Clean up the commented out code and make sure the comments are unified. Carrying around the commented out code means people making changes to model_cards are supposed to update it, but that's not clear and won't be picked up by type checking etc. Drop it for now - it's in the git history. Also make the rest of the comments a bit more uniform, and place comments about a specific model card inside the model card (instead of above) so they don't get lost when code is added/moved around. Test plan: - my eyes	2026-01-15 13:21:58 +00:00
Evan Quiney	82ba42bae9	add glm-47, minimax-m21 (#1147 ) Adds support glm 4.7 and MiniMax M2.1 Manual testing: Tensor + Pipeline execution of both models. Closes #1141 and #1142	2026-01-14 16:33:17 +00:00
Jake Hillion	3671528fa4	nix: add dashboard build with dream2nix Continue working towards a fully Nix based build by building the dashboard with Nix. Continuing the theme of using the existing lock files, use dream2nix to parse the lock file and build the tree of dependency derivations. dream2nix doesn't like the bundleDependencies, so we apply a small patch to the lock file that drops all dependencies that are bundled. This should ideally be contributed upstream but that can be done later. Use this new dashboard build in the build-app CI workflow, meaning future macOS apps will include this reproducible dashboard. Test plan: - Built a DMG, shipped to a cluster, loaded in a browser with no cache and the dashboard looks good. - Directory layout is as expected: ``` $ nix build .#dashboard $ find result/ ... result/_app/immutable/entry result/_app/immutable/entry/app.CTPAnMjf.js result/_app/immutable/entry/start.fUSEa-2O.js result/_app/immutable/nodes result/_app/immutable/nodes/3.DqQr1Obm.js result/_app/immutable/nodes/0.DgEY44RO.js result/_app/immutable/nodes/2.BjZg_lJh.js result/_app/immutable/nodes/1.D6vGUYYT.js result/_app/env.js result/_app/version.json result/exo-logo.png result/favicon.ico result/index.html ```	2026-01-14 15:58:16 +01:00
Jake Hillion	e6434ec446	nix: add Rust builds with crane and fenix The Rust workspace lacked Nix build support, making it difficult to build packages reproducibly or run checks in CI. Added a flake-parts module at rust/parts.nix that uses crane for Rust builds and fenix for the nightly toolchain. The source filter isolates rust/ and root Cargo files to prevent Python/docs changes from triggering Rust rebuilds. Exports packages (system_custodian, exo_pyo3_bindings wheel, exo-rust-workspace) and checks (cargo-nextest, cargo-doc) for all three target platforms. The devShell now uses inputsFrom to inherit build dependencies from the workspace package, removing the need for manual pkg-config/openssl setup. Test plan: - Ran `nix flake check` successfully - Built `nix build ".#checks.x86_64-linux.cargo-nextest"` and tests pass - Built `nix build ".#exo_pyo3_bindings"` and wheel is produced	2026-01-14 11:52:29 +00:00
Jake Hillion	bdb43e1dbb	nix: drop noisy echos from devshell Drop all the printing when entering a devshell. It's annoying, and not a super accurate description of how to develop exo anyway.	2026-01-14 10:04:57 +00:00
Jake Hillion	e4a01e2b0e	chore(deps): nix lock file maintenance Update nix flake inputs. Add a second input as Swift is currently broken in nixpkgs on Linux for `swift-format` as we want `nix fmt` to continue being reproducible everywhere.	2026-01-13 19:57:14 +01:00
Evan Quiney	1200a7db64	Add tensor sharding for GPT-OSS (#1144 ) ## Motivation GPT OSS did not previously support tensor sharding ## Changes Add GPT sharding support in tensor_auto_parallel. Code is mostly @rltakashige's ## Test Plan ### Manual Testing Tested GPT-OSS - MLX Fast Sync causes issues in Tensor RDMA - this is a general problem at the moment.	2026-01-13 17:25:52 +00:00