diff --git a/dev/model_user_guide.ipynb b/dev/model_user_guide.ipynb
index df849af..4ae58f8 100644
--- a/dev/model_user_guide.ipynb
+++ b/dev/model_user_guide.ipynb
@@ -77,7 +77,6 @@
     "import os\n",
     "import tempfile\n",
     "from collections.abc import Sequence\n",
-    "from typing import Optional\n",
     "\n",
     "import numpy as np\n",
     "import scvi\n",
@@ -294,10 +293,10 @@
     "    def setup_anndata(\n",
     "        cls,\n",
     "        adata: AnnData,\n",
-    "        batch_key: Optional[str] = None,\n",
-    "        layer: Optional[str] = None,\n",
+    "        batch_key: str | None = None,\n",
+    "        layer: str | None = None,\n",
     "        **kwargs,\n",
-    "    ) -> Optional[AnnData]:\n",
+    "    ) -> AnnData | None:\n",
     "        setup_method_args = cls._get_setup_method_args(**locals())\n",
     "        anndata_fields = [\n",
     "            LayerField(REGISTRY_KEYS.X_KEY, layer, is_count_data=True),\n",
@@ -1257,9 +1256,9 @@
     "@torch.inference_mode()\n",
     "def get_latent_representation(\n",
     "    self,\n",
-    "    adata: Optional[AnnData] = None,\n",
-    "    indices: Optional[Sequence[int]] = None,\n",
-    "    batch_size: Optional[int] = None,\n",
+    "    adata: AnnData | None = None,\n",
+    "    indices: Sequence[int] | None = None,\n",
+    "    batch_size: int | None = None,\n",
     ") -> np.ndarray:\n",
     "    r\"\"\"Return the latent representation for each cell.\n",
     "\n",
@@ -1466,10 +1465,10 @@
     "    def setup_anndata(\n",
     "        cls,\n",
     "        adata: AnnData,\n",
-    "        batch_key: Optional[str] = None,\n",
-    "        layer: Optional[str] = None,\n",
+    "        batch_key: str | None = None,\n",
+    "        layer: str | None = None,\n",
     "        **kwargs,\n",
-    "    ) -> Optional[AnnData]:\n",
+    "    ) -> AnnData | None:\n",
     "        setup_method_args = cls._get_setup_method_args(**locals())\n",
     "        anndata_fields = [\n",
     "            LayerField(REGISTRY_KEYS.X_KEY, layer, is_count_data=True),\n",
diff --git a/scrna/AutoZI_tutorial.ipynb b/scrna/AutoZI_tutorial.ipynb
index 30f9224..8738f3f 100644
--- a/scrna/AutoZI_tutorial.ipynb
+++ b/scrna/AutoZI_tutorial.ipynb
@@ -10590,7 +10590,7 @@
     "ct = adata.obs.str_labels.astype(\"category\")\n",
     "codes = np.unique(ct.cat.codes)\n",
     "cats = ct.cat.categories\n",
-    "for ind_cell_type, cell_type in zip(codes, cats):\n",
+    "for ind_cell_type, cell_type in zip(codes, cats, strict=False):\n",
     "    is_zi_pred_genelabel_here = is_zi_pred_genelabel[:, ind_cell_type]\n",
     "    print(\n",
     "        f\"Fraction of predicted ZI genes for cell type {cell_type} :\",\n",
@@ -10654,7 +10654,7 @@
    ],
    "source": [
     "# With avg expressions > 1\n",
-    "for ind_cell_type, cell_type in zip(codes, cats):\n",
+    "for ind_cell_type, cell_type in zip(codes, cats, strict=False):\n",
     "    mask_sufficient_expression = (\n",
     "        np.array(adata.X[adata.obs.str_labels.values.reshape(-1) == cell_type, :].mean(axis=0))\n",
     "        > 1.0\n",
diff --git a/scrna/scanvi_fix.ipynb b/scrna/scanvi_fix.ipynb
index 1c94d65..7ad0794 100644
--- a/scrna/scanvi_fix.ipynb
+++ b/scrna/scanvi_fix.ipynb
@@ -15232,8 +15232,8 @@
     "    models = [model_no_fix, model_fix, model_fix_linear]\n",
     "    model_names = [\"No fix\", \"Fix\", \"Fix linear\"]\n",
     "\n",
-    "    for i, (metric, ylim) in enumerate(zip(metrics, ylims)):\n",
-    "        for j, (model, model_name) in enumerate(zip(models, model_names)):\n",
+    "    for i, (metric, ylim) in enumerate(zip(metrics, ylims, strict=False)):\n",
+    "        for j, (model, model_name) in enumerate(zip(models, model_names, strict=False)):\n",
     "            plot_metric(axes[i, j], metric, model, model_name, ylim=ylim)\n",
     "\n",
     "    fig.text(-0.01, 0.8, \"Classification loss\", va=\"center\", rotation=\"vertical\")\n",
@@ -15343,7 +15343,7 @@
     "    models = [model_no_fix, model_fix, model_fix_linear]\n",
     "    model_names = [\"No fix\", \"Fix\", \"Fix linear\"]\n",
     "\n",
-    "    for model, model_name, ax in zip(models, model_names, axes):\n",
+    "    for model, model_name, ax in zip(models, model_names, axes, strict=False):\n",
     "        plot_confusion_matrix(ax, model, model_name, subset)\n",
     "\n",
     "    fig.text(0.0, 0.5, \"Observed\", va=\"center\", rotation=\"vertical\")\n",
@@ -15492,7 +15492,7 @@
     "    model_names = [\"No fix\", \"Fix\", \"Fix linear\"]\n",
     "    legend_loc = [\"none\", \"none\", \"right margin\"]\n",
     "\n",
-    "    for model, model_name, ax, leg_loc in zip(models, model_names, axes, legend_loc):\n",
+    "    for model, model_name, ax, leg_loc in zip(models, model_names, axes, legend_loc, strict=False):\n",
     "        plot_latent_mde(ax, model, model_name, subset, leg_loc)\n",
     "\n",
     "    fig.text(0.0, 0.5, \"MDE_2\", va=\"center\", rotation=\"vertical\")\n",