diff --git a/conda-envs/environment-dev.yml b/conda-envs/environment-dev.yml
index 231dfa05cf..230dfcb62f 100644
--- a/conda-envs/environment-dev.yml
+++ b/conda-envs/environment-dev.yml
@@ -6,6 +6,7 @@ channels:
 dependencies:
 # Base dependencies
 - arviz>=0.13.0
+- arviz-base
 - blas
 - cachetools>=4.2.1
 - cloudpickle
diff --git a/conda-envs/environment-docs.yml b/conda-envs/environment-docs.yml
index f85f8fc55b..e8ea8cbd1a 100644
--- a/conda-envs/environment-docs.yml
+++ b/conda-envs/environment-docs.yml
@@ -6,6 +6,7 @@ channels:
 dependencies:
 # Base dependencies
 - arviz>=0.13.0
+- arviz-base
 - cachetools>=4.2.1
 - cloudpickle
 - numpy>=1.25.0
diff --git a/pymc/backends/arviz.py b/pymc/backends/arviz.py
index 63f8370523..4b4ede3935 100644
--- a/pymc/backends/arviz.py
+++ b/pymc/backends/arviz.py
@@ -29,7 +29,8 @@
 import xarray
 
 from arviz import InferenceData, concat, rcParams
-from arviz.data.base import CoordSpec, DimSpec, dict_to_dataset, requires
+from arviz.data.base import CoordSpec, DimSpec, requires
+from arviz_base import dict_to_dataset
 from pytensor.graph import ancestors
 from pytensor.tensor.sharedvar import SharedVariable
 from rich.progress import Console
@@ -305,14 +306,14 @@ def posterior_to_xarray(self):
         return (
             dict_to_dataset(
                 data,
-                library=pymc,
+                inference_library=pymc,
                 coords=self.coords,
                 dims=self.dims,
                 attrs=self.attrs,
             ),
             dict_to_dataset(
                 data_warmup,
-                library=pymc,
+                inference_library=pymc,
                 coords=self.coords,
                 dims=self.dims,
                 attrs=self.attrs,
@@ -347,14 +348,14 @@ def sample_stats_to_xarray(self):
         return (
             dict_to_dataset(
                 data,
-                library=pymc,
+                inference_library=pymc,
                 dims=None,
                 coords=self.coords,
                 attrs=self.attrs,
             ),
             dict_to_dataset(
                 data_warmup,
-                library=pymc,
+                inference_library=pymc,
                 dims=None,
                 coords=self.coords,
                 attrs=self.attrs,
@@ -367,7 +368,11 @@ def posterior_predictive_to_xarray(self):
         data = self.posterior_predictive
         dims = {var_name: self.sample_dims + self.dims.get(var_name, []) for var_name in data}
         return dict_to_dataset(
-            data, library=pymc, coords=self.coords, dims=dims, default_dims=self.sample_dims
+            data,
+            inference_library=pymc,
+            coords=self.coords,
+            dims=dims,
+            sample_dims=self.sample_dims,
         )
 
     @requires(["predictions"])
@@ -376,7 +381,11 @@ def predictions_to_xarray(self):
         data = self.predictions
         dims = {var_name: self.sample_dims + self.dims.get(var_name, []) for var_name in data}
         return dict_to_dataset(
-            data, library=pymc, coords=self.coords, dims=dims, default_dims=self.sample_dims
+            data,
+            inference_library=pymc,
+            coords=self.coords,
+            dims=dims,
+            sample_dims=self.sample_dims,
         )
 
     def priors_to_xarray(self):
@@ -399,7 +408,7 @@ def priors_to_xarray(self):
                 if var_names is None
                 else dict_to_dataset_drop_incompatible_coords(
                     {k: np.expand_dims(self.prior[k], 0) for k in var_names},
-                    library=pymc,
+                    inference_library=pymc,
                     coords=self.coords,
                     dims=self.dims,
                 )
@@ -414,10 +423,10 @@ def observed_data_to_xarray(self):
             return None
         return dict_to_dataset(
             self.observations,
-            library=pymc,
+            inference_library=pymc,
             coords=self.coords,
             dims=self.dims,
-            default_dims=[],
+            sample_dims=[],
         )
 
     @requires("model")
@@ -429,10 +438,10 @@ def constant_data_to_xarray(self):
 
         xarray_dataset = dict_to_dataset(
             constant_data,
-            library=pymc,
+            inference_library=pymc,
             coords=self.coords,
             dims=self.dims,
-            default_dims=[],
+            sample_dims=[],
         )
 
         # provisional handling of scalars in constant
@@ -707,9 +716,9 @@ def apply_function_over_dataset(
 
     return dict_to_dataset(
         out_trace,
-        library=pymc,
+        inference_library=pymc,
         dims=dims,
         coords=coords,
-        default_dims=list(sample_dims),
+        sample_dims=list(sample_dims),
         skip_event_dims=True,
     )
diff --git a/pymc/smc/sampling.py b/pymc/smc/sampling.py
index 5afd398281..352224026e 100644
--- a/pymc/smc/sampling.py
+++ b/pymc/smc/sampling.py
@@ -267,7 +267,9 @@ def _save_sample_stats(
         sample_stats = dict_to_dataset(
             sample_stats_dict,
             attrs=sample_settings_dict,
-            library=pymc,
+            inference_library=pymc,
+            sample_dims=["chain"],
+            check_conventions=False,
         )
 
         ikwargs: dict[str, Any] = {"model": model}
diff --git a/requirements-dev.txt b/requirements-dev.txt
index 22bcdaf9ea..25bc83223e 100644
--- a/requirements-dev.txt
+++ b/requirements-dev.txt
@@ -1,6 +1,7 @@
 # This file is auto-generated by scripts/generate_pip_deps_from_conda.py, do not modify.
 # See that file for comments about the need/usage of each dependency.
 
+arviz-base
 arviz>=0.13.0
 cachetools>=4.2.1
 cloudpickle