diff --git a/.gitignore b/.gitignore
index be92b9c482..28e08b3686 100644
--- a/.gitignore
+++ b/.gitignore
@@ -32,4 +32,4 @@ log/
 
 
 # docs autogen
-/docs/source/api_docs/*/
+/docs/source/api_docs/doc_gen/
diff --git a/daft/__init__.py b/daft/__init__.py
index 8ad5025a18..e18ac11455 100644
--- a/daft/__init__.py
+++ b/daft/__init__.py
@@ -77,7 +77,7 @@ class daft:
     from_ray_dataset,
 )
 from daft.dataframe import DataFrame
-from daft.datatype import DataType
+from daft.datatype import DataType, ImageMode
 from daft.expressions import col, lit
 from daft.io import from_glob_path, read_csv, read_json, read_parquet
 from daft.series import Series
@@ -98,6 +98,7 @@ class daft:
     "DataFrame",
     "col",
     "DataType",
+    "ImageMode",
     "lit",
     "Series",
     "register_viz_hook",
diff --git a/daft/daft.pyi b/daft/daft.pyi
index 159c3bf7c6..739171cc77 100644
--- a/daft/daft.pyi
+++ b/daft/daft.pyi
@@ -6,31 +6,45 @@ class ImageMode(Enum):
     Supported image modes for Daft's image type.
     """
 
-    # 8-bit grayscale
+    #: 8-bit grayscale
     L: int
-    # 8-bit grayscale + alpha
+
+    #: 8-bit grayscale + alpha
     LA: int
-    # 8-bit RGB
+
+    #: 8-bit RGB
     RGB: int
-    # 8-bit RGB + alpha
+
+    #: 8-bit RGB + alpha
     RGBA: int
-    # 16-bit grayscale
+
+    #: 16-bit grayscale
     L16: int
-    # 16-bit grayscale + alpha
+
+    #: 16-bit grayscale + alpha
     LA16: int
-    # 16-bit RGB
+
+    #: 16-bit RGB
     RGB16: int
-    # 16-bit RGB + alpha
+
+    #: 16-bit RGB + alpha
     RGBA16: int
-    # 32-bit floating RGB
+
+    #: 32-bit floating RGB
     RGB32F: int
-    # 32-bit floating RGB + alpha
+
+    #: 32-bit floating RGB + alpha
     RGBA32F: int
 
     @staticmethod
     def from_mode_string(mode: str) -> ImageMode:
         """
         Create an ImageMode from its string representation.
+
+        Args:
+            mode: String representation of the mode. This is the same as the enum
+                attribute name, e.g. ``ImageMode.from_mode_string("RGB")`` would
+                return ``ImageMode.RGB``.
         """
         ...
 
diff --git a/daft/datatype.py b/daft/datatype.py
index 539e0c20d0..b682d4084b 100644
--- a/daft/datatype.py
+++ b/daft/datatype.py
@@ -176,6 +176,28 @@ def embedding(cls, name: str, dtype: DataType, size: int) -> DataType:
     def image(
         cls, mode: str | ImageMode | None = None, height: int | None = None, width: int | None = None
     ) -> DataType:
+        """Create an Image DataType: image arrays contain (height, width, channel) ndarrays of pixel values.
+
+        Each image in the array has an :class:`~daft.ImageMode`, which describes the pixel dtype (e.g. uint8) and
+        the number of image channels/bands and their logical interpretation (e.g. RGB).
+
+        If the height, width, and mode are the same for all images in the array, specifying them when constructing
+        this type is advised, since that will allow Daft to create a more optimized physical representation
+        of the image array.
+
+        If the height, width, or mode may vary across images in the array, leaving these fields unspecified when
+        creating this type will cause Daft to respresent this image array as a heterogeneous collection of images,
+        where each image can have a different mode, height, and width. This is much more flexible, but will result
+        in a less compact representation and may be make some operations less efficient.
+
+        Args:
+            mode: The mode of the image. By default, this is inferred from the underlying data.
+                If height and width are specified, the mode must also be specified.
+            height: The height of the image. By default, this is inferred from the underlying data.
+                Must be specified if the width is specified.
+            width: The width of the image. By default, this is inferred from the underlying data.
+                Must be specified if the width is specified.
+        """
         if isinstance(mode, str):
             mode = ImageMode.from_mode_string(mode)
         if height is not None and width is not None:
diff --git a/docs/source/api_docs/context.rst b/docs/source/api_docs/context.rst
index 5c10a46f9a..e7d94488c8 100644
--- a/docs/source/api_docs/context.rst
+++ b/docs/source/api_docs/context.rst
@@ -8,7 +8,7 @@ Control the execution backend that Daft will run on by calling these functions o
 
 .. autosummary::
     :nosignatures:
-    :toctree: configuration_functions
+    :toctree: doc_gen/configuration_functions
 
     daft.context.set_runner_py
     daft.context.set_runner_ray
diff --git a/docs/source/api_docs/dataframe.rst b/docs/source/api_docs/dataframe.rst
index 0b72f1e229..5c7145aca4 100644
--- a/docs/source/api_docs/dataframe.rst
+++ b/docs/source/api_docs/dataframe.rst
@@ -5,7 +5,7 @@ DataFrame
 
 .. autosummary::
     :nosignatures:
-    :toctree: dataframe_methods
+    :toctree: doc_gen/dataframe_methods
 
     DataFrame
 
@@ -26,7 +26,7 @@ Manipulating Columns
 
 .. autosummary::
     :nosignatures:
-    :toctree: dataframe_methods
+    :toctree: doc_gen/dataframe_methods
 
     daft.DataFrame.select
     daft.DataFrame.with_column
@@ -41,7 +41,7 @@ Filtering Rows
 
 .. autosummary::
     :nosignatures:
-    :toctree: dataframe_methods
+    :toctree: doc_gen/dataframe_methods
 
     daft.DataFrame.distinct
     daft.DataFrame.where
@@ -54,7 +54,7 @@ Reordering
 
 .. autosummary::
     :nosignatures:
-    :toctree: dataframe_methods
+    :toctree: doc_gen/dataframe_methods
 
     daft.DataFrame.sort
     daft.DataFrame.repartition
@@ -66,7 +66,7 @@ Combining
 
 .. autosummary::
     :nosignatures:
-    :toctree: dataframe_methods
+    :toctree: doc_gen/dataframe_methods
 
     daft.DataFrame.join
 
@@ -81,7 +81,7 @@ Aggregations
 
 .. autosummary::
     :nosignatures:
-    :toctree: dataframe_methods
+    :toctree: doc_gen/dataframe_methods
 
     daft.DataFrame.groupby
     daft.DataFrame.sum
@@ -102,7 +102,7 @@ Materialization
 
 .. autosummary::
     :nosignatures:
-    :toctree: dataframe_methods
+    :toctree: doc_gen/dataframe_methods
 
     daft.DataFrame.collect
 
@@ -113,7 +113,7 @@ Visualization
 
 .. autosummary::
     :nosignatures:
-    :toctree: dataframe_methods
+    :toctree: doc_gen/dataframe_methods
 
     daft.DataFrame.show
 
@@ -127,7 +127,7 @@ Writing Data
 
 .. autosummary::
     :nosignatures:
-    :toctree: dataframe_methods
+    :toctree: doc_gen/dataframe_methods
 
     daft.DataFrame.write_parquet
     daft.DataFrame.write_csv
@@ -139,7 +139,7 @@ Integrations
 
 .. autosummary::
     :nosignatures:
-    :toctree: dataframe_methods
+    :toctree: doc_gen/dataframe_methods
 
     daft.DataFrame.to_pandas
     daft.DataFrame.to_ray_dataset
@@ -150,7 +150,7 @@ Schema and Lineage
 
 .. autosummary::
     :nosignatures:
-    :toctree: dataframe_methods
+    :toctree: doc_gen/dataframe_methods
 
     daft.DataFrame.explain
     daft.DataFrame.schema
diff --git a/docs/source/api_docs/datatype.rst b/docs/source/api_docs/datatype.rst
index 33a498c350..bd4cf3e1cc 100644
--- a/docs/source/api_docs/datatype.rst
+++ b/docs/source/api_docs/datatype.rst
@@ -5,7 +5,7 @@ DataTypes
 
 .. autosummary::
     :nosignatures:
-    :toctree: datatype_methods
+    :toctree: doc_gen/datatype_methods
 
     daft.DataType
 
@@ -30,7 +30,7 @@ Numeric
 
 .. autosummary::
     :nosignatures:
-    :toctree: datatype_methods
+    :toctree: doc_gen/datatype_methods
 
     daft.DataType.int8
     daft.DataType.int16
@@ -51,7 +51,7 @@ Logical
 
 .. autosummary::
     :nosignatures:
-    :toctree: datatype_methods
+    :toctree: doc_gen/datatype_methods
 
     daft.DataType.bool
 
@@ -63,7 +63,7 @@ Strings
 
 .. autosummary::
     :nosignatures:
-    :toctree: datatype_methods
+    :toctree: doc_gen/datatype_methods
 
     daft.DataType.binary
     daft.DataType.string
@@ -76,7 +76,7 @@ Temporal
 
 .. autosummary::
     :nosignatures:
-    :toctree: datatype_methods
+    :toctree: doc_gen/datatype_methods
 
     daft.DataType.date
 
@@ -88,7 +88,7 @@ Nested
 
 .. autosummary::
     :nosignatures:
-    :toctree: datatype_methods
+    :toctree: doc_gen/datatype_methods
 
     daft.DataType.list
     daft.DataType.fixed_size_list
@@ -100,7 +100,7 @@ Python
 
 .. autosummary::
     :nosignatures:
-    :toctree: datatype_methods
+    :toctree: doc_gen/datatype_methods
 
     daft.DataType.python
 
@@ -115,7 +115,7 @@ Machine Learning
 
 .. autosummary::
     :nosignatures:
-    :toctree: datatype_methods
+    :toctree: doc_gen/datatype_methods
 
     daft.DataType.embedding
 
@@ -124,15 +124,25 @@ Computer Vision
 
 .. autosummary::
     :nosignatures:
-    :toctree: datatype_methods
+    :toctree: doc_gen/datatype_methods
 
     daft.DataType.image
 
+.. autosummary::
+    :nosignatures:
+
+    ImageMode
+
 
 Miscellaneous
 ^^^^^^^^^^^^^
 .. autosummary::
     :nosignatures:
-    :toctree: datatype_methods
+    :toctree: doc_gen/datatype_methods
 
     daft.DataType.null
+
+.. toctree::
+    :hidden:
+
+    datatype_image_mode/daft.ImageMode
diff --git a/docs/source/api_docs/datatype_image_mode/daft.ImageMode.from_mode_string.rst b/docs/source/api_docs/datatype_image_mode/daft.ImageMode.from_mode_string.rst
new file mode 100644
index 0000000000..c0b9a4392f
--- /dev/null
+++ b/docs/source/api_docs/datatype_image_mode/daft.ImageMode.from_mode_string.rst
@@ -0,0 +1,6 @@
+daft.ImageMode.from\_mode\_string
+=================================
+
+.. currentmodule:: daft
+
+.. automethod:: ImageMode.from_mode_string
\ No newline at end of file
diff --git a/docs/source/api_docs/datatype_image_mode/daft.ImageMode.rst b/docs/source/api_docs/datatype_image_mode/daft.ImageMode.rst
new file mode 100644
index 0000000000..6699b6f221
--- /dev/null
+++ b/docs/source/api_docs/datatype_image_mode/daft.ImageMode.rst
@@ -0,0 +1,27 @@
+daft.ImageMode
+==============
+
+.. currentmodule:: daft
+
+.. autoclass:: ImageMode
+
+   .. autosummary::
+      :toctree:
+
+      ~ImageMode.from_mode_string
+
+   .. rubric:: Variants
+
+   .. autosummary::
+   
+      ~ImageMode.L
+      ~ImageMode.LA
+      ~ImageMode.RGB
+      ~ImageMode.RGBA
+      ~ImageMode.L16
+      ~ImageMode.LA16
+      ~ImageMode.RGB16
+      ~ImageMode.RGBA16
+      ~ImageMode.RGB32F
+      ~ImageMode.RGBA32F
+   
\ No newline at end of file
diff --git a/docs/source/api_docs/expressions.rst b/docs/source/api_docs/expressions.rst
index ff2ab700c0..a61825a5f1 100644
--- a/docs/source/api_docs/expressions.rst
+++ b/docs/source/api_docs/expressions.rst
@@ -5,7 +5,7 @@ Expressions
 
 .. autosummary::
     :nosignatures:
-    :toctree: expression_methods
+    :toctree: doc_gen/expression_methods
 
     daft.expressions.Expression
 
@@ -14,7 +14,7 @@ Expression Constructors
 
 .. autosummary::
     :nosignatures:
-    :toctree: expression_methods
+    :toctree: doc_gen/expression_methods
 
     daft.DataFrame.__getitem__
     daft.expressions.col
@@ -31,7 +31,7 @@ Numeric
 Operations on numbers (floats and integers)
 
 .. autosummary::
-    :toctree: expression_methods
+    :toctree: doc_gen/expression_methods
 
     daft.expressions.Expression.__abs__
     daft.expressions.Expression.__add__
@@ -46,7 +46,7 @@ Logical
 Operations on logical expressions (True/False booleans)
 
 .. autosummary::
-    :toctree: expression_methods
+    :toctree: doc_gen/expression_methods
 
     daft.expressions.Expression.__invert__
     daft.expressions.Expression.__and__
@@ -61,7 +61,7 @@ Comparisons
 Comparing expressions and values, returning a logical expression
 
 .. autosummary::
-    :toctree: expression_methods
+    :toctree: doc_gen/expression_methods
 
     daft.expressions.Expression.__lt__
     daft.expressions.Expression.__le__
@@ -83,7 +83,7 @@ Operations on strings, accessible through the ``Expression.float`` method access
 Example: ``e1.float.is_nan()``
 
 .. autosummary::
-    :toctree: expression_methods
+    :toctree: doc_gen/expression_methods
 
     daft.expressions.expressions.ExpressionFloatNamespace.is_nan
 
@@ -97,7 +97,7 @@ Operations on strings, accessible through the ``Expression.str`` method accessor
 Example: ``e1.str.concat(e2)``
 
 .. autosummary::
-    :toctree: expression_methods
+    :toctree: doc_gen/expression_methods
 
     daft.expressions.expressions.ExpressionStringNamespace.concat
     daft.expressions.expressions.ExpressionStringNamespace.contains
@@ -116,7 +116,7 @@ Example: ``e.dt.day()``
 
 .. autosummary::
     :nosignatures:
-    :toctree: expression_methods
+    :toctree: doc_gen/expression_methods
 
     daft.expressions.expressions.ExpressionDatetimeNamespace.day
     daft.expressions.expressions.ExpressionDatetimeNamespace.month
@@ -134,7 +134,7 @@ Example: ``e.url.download()``
 
 .. autosummary::
     :nosignatures:
-    :toctree: expression_methods
+    :toctree: doc_gen/expression_methods
 
     daft.expressions.expressions.ExpressionUrlNamespace.download
 
@@ -149,7 +149,7 @@ Example: ``e.image.resize()``
 
 .. autosummary::
     :nosignatures:
-    :toctree: expression_methods
+    :toctree: doc_gen/expression_methods
 
     daft.expressions.expressions.ExpressionImageNamespace.resize
     daft.expressions.expressions.ExpressionImageNamespace.decode
@@ -160,7 +160,7 @@ Changing Column Names/Types
 
 .. autosummary::
     :nosignatures:
-    :toctree: expression_methods
+    :toctree: doc_gen/expression_methods
 
     daft.expressions.Expression.alias
     daft.expressions.Expression.cast
@@ -170,6 +170,6 @@ Running Python Functions
 
 .. autosummary::
     :nosignatures:
-    :toctree: expression_methods
+    :toctree: doc_gen/expression_methods
 
     daft.expressions.Expression.apply
diff --git a/docs/source/api_docs/groupby.rst b/docs/source/api_docs/groupby.rst
index 2e0513165f..342132e218 100644
--- a/docs/source/api_docs/groupby.rst
+++ b/docs/source/api_docs/groupby.rst
@@ -9,7 +9,7 @@ Calling ``DataFrame.groupby`` returns a ``GroupedDataFrame`` object which is a v
 
 .. autosummary::
     :nosignatures:
-    :toctree: dataframe_methods
+    :toctree: doc_gen/dataframe_methods
 
     daft.dataframe.dataframe.GroupedDataFrame.sum
     daft.dataframe.dataframe.GroupedDataFrame.mean
diff --git a/docs/source/api_docs/input_output.rst b/docs/source/api_docs/input_output.rst
index df5ba0cd75..cec6ddb678 100644
--- a/docs/source/api_docs/input_output.rst
+++ b/docs/source/api_docs/input_output.rst
@@ -15,7 +15,7 @@ Python Objects
 
 .. autosummary::
     :nosignatures:
-    :toctree: io_functions
+    :toctree: doc_gen/io_functions
 
     daft.from_pylist
     daft.from_pydict
@@ -26,12 +26,12 @@ Arrow
 
 .. autosummary::
     :nosignatures:
-    :toctree: io_functions
+    :toctree: doc_gen/io_functions
 
 
 .. autosummary::
     :nosignatures:
-    :toctree: io_functions
+    :toctree: doc_gen/io_functions
 
     daft.from_arrow
     daft.DataFrame.to_arrow
@@ -41,7 +41,7 @@ Pandas
 
 .. autosummary::
     :nosignatures:
-    :toctree: io_functions
+    :toctree: doc_gen/io_functions
 
     daft.from_pandas
     daft.DataFrame.to_pandas
@@ -51,7 +51,7 @@ File Paths
 
 .. autosummary::
     :nosignatures:
-    :toctree: io_functions
+    :toctree: doc_gen/io_functions
 
     daft.from_glob_path
 
@@ -65,7 +65,7 @@ Parquet
 
 .. autosummary::
     :nosignatures:
-    :toctree: io_functions
+    :toctree: doc_gen/io_functions
 
     daft.read_parquet
     daft.DataFrame.write_parquet
@@ -75,7 +75,7 @@ CSV
 
 .. autosummary::
     :nosignatures:
-    :toctree: io_functions
+    :toctree: doc_gen/io_functions
 
     daft.read_csv
     daft.DataFrame.write_csv
@@ -85,7 +85,7 @@ JSON
 
 .. autosummary::
     :nosignatures:
-    :toctree: io_functions
+    :toctree: doc_gen/io_functions
 
     daft.read_json
 
@@ -99,7 +99,7 @@ Ray Datasets
 
 .. autosummary::
     :nosignatures:
-    :toctree: io_functions
+    :toctree: doc_gen/io_functions
 
     daft.from_ray_dataset
     daft.DataFrame.to_ray_dataset
@@ -109,7 +109,7 @@ Dask
 
 .. autosummary::
     :nosignatures:
-    :toctree: io_functions
+    :toctree: doc_gen/io_functions
 
     daft.from_dask_dataframe
     daft.DataFrame.to_dask_dataframe
diff --git a/docs/source/learn/user_guides/datatypes.rst b/docs/source/learn/user_guides/datatypes.rst
index 3177edd051..3f0e163e60 100644
--- a/docs/source/learn/user_guides/datatypes.rst
+++ b/docs/source/learn/user_guides/datatypes.rst
@@ -111,8 +111,8 @@ Complex Types
 
 Daft supports many more interesting complex DataTypes, for example:
 
-* :meth:`~daft.DataType.embedding()`: Lower-dimensional vector representation of data (e.g. words)
-* :meth:`~daft.DataType.image()`: NHWC images
+* :meth:`DataType.embedding() <daft.DataType.embedding>`: Lower-dimensional vector representation of data (e.g. words)
+* :meth:`DataType.image() <daft.DataType.image>`: NHWC images
 
 Daft abstracts away the in-memory representation of your data and provides kernels for many common operations on top of these data types. For supported image operations see the :ref:`image expressions API reference <api-expressions-images>`.
 
diff --git a/src/datatypes/image_mode.rs b/src/datatypes/image_mode.rs
index 8a039d002c..d0e8c5331c 100644
--- a/src/datatypes/image_mode.rs
+++ b/src/datatypes/image_mode.rs
@@ -13,16 +13,16 @@ use crate::{
 
 /// Supported image modes for Daft's image type.
 ///
-/// L       - 8-bit grayscale
-/// LA      - 8-bit grayscale + alpha
-/// RGB     - 8-bit RGB
-/// RGBA    - 8-bit RGB + alpha
-/// L16     - 16-bit grayscale
-/// LA16    - 16-bit grayscale + alpha
-/// RGB16   - 16-bit RGB
-/// RGBA16  - 16-bit RGB + alpha
-/// RGB32F  - 32-bit floating RGB
-/// RGBA32F - 32-bit floating RGB + alpha
+/// | L       - 8-bit grayscale
+/// | LA      - 8-bit grayscale + alpha
+/// | RGB     - 8-bit RGB
+/// | RGBA    - 8-bit RGB + alpha
+/// | L16     - 16-bit grayscale
+/// | LA16    - 16-bit grayscale + alpha
+/// | RGB16   - 16-bit RGB
+/// | RGBA16  - 16-bit RGB + alpha
+/// | RGB32F  - 32-bit floating RGB
+/// | RGBA32F - 32-bit floating RGB + alpha
 #[allow(clippy::upper_case_acronyms)]
 #[derive(Clone, Copy, Debug, PartialEq, Eq, Serialize, Deserialize, Hash, FromPrimitive)]
 #[cfg_attr(feature = "python", pyclass)]
@@ -43,6 +43,11 @@ pub enum ImageMode {
 #[pymethods]
 impl ImageMode {
     /// Create an ImageMode from its string representation.
+    ///
+    /// Args:
+    ///     mode: String representation of the mode. This is the same as the enum
+    ///         attribute name, e.g. ``ImageMode.from_mode_string("RGB")`` would
+    ///         return ``ImageMode.RGB``.
     #[staticmethod]
     pub fn from_mode_string(mode: &str) -> PyResult<Self> {
         Self::from_str(mode).map_err(|e| PyValueError::new_err(e.to_string()))