diff --git a/tests/unit/vertexai/test_prompt.py b/tests/unit/vertexai/test_prompt.py
new file mode 100644
index 00000000000..e59c6ca4385
--- /dev/null
+++ b/tests/unit/vertexai/test_prompt.py
@@ -0,0 +1,186 @@
+# -*- coding: utf-8 -*-
+
+# Copyright 2024 Google LLC
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+#
+"""Unit tests for generative model prompts."""
+# pylint: disable=protected-access,bad-continuation
+
+from vertexai.generative_models._prompt import Prompt
+from vertexai.generative_models import Content, Part
+
+import pytest
+
+from typing import Any
+
+
+def _is_list_type(obj: Any, T: Any) -> bool:
+    return isinstance(obj, list) and all(isinstance(s, T) for s in obj)
+
+
+@pytest.mark.usefixtures("google_auth_mock")
+class TestPrompt:
+    """Unit tests for generative model prompts."""
+
+    def test_inferred_prompt_constructor(self):
+        # Create inferred prompt with string only variable values
+        prompt = Prompt(
+            prompt_data="Rate the movie {movie1}",
+            variables={
+                "movie1": "The Avengers",
+            },
+        )
+        # Inferred prompt data should remain as string before compilation
+        assert prompt.prompt_data == "Rate the movie {movie1}"
+        # Variables values should be converted to List[Part]
+        assert _is_list_type(prompt.variables["movie1"], Part)
+
+        # Create inferred prompt with List[Part] variable values
+        prompt = Prompt(
+            prompt_data="Rate the movie {movie1}",
+            variables={
+                "movie1": [Part.from_text("The Avengers")],
+            },
+        )
+        # Variables values should be converted to List[Part]
+        assert _is_list_type(prompt.variables["movie1"], Part)
+
+        # Inferred prompt variables must either be string or List[Part]
+        with pytest.raises(ValueError):
+            Prompt(
+                prompt_data="Rate the movie {movie1}",
+                variables={
+                    "movie1": Part.from_text("The Avengers"),
+                },
+            )
+
+    def test_inferred_prompt_to_content(self):
+        prompt = Prompt(
+            prompt_data="Which movie is better, {movie1} or {movie2}?",
+            variables={
+                "movie1": "The Avengers",
+                "movie2": "Frozen",
+            },
+        )
+        unassembled_prompt_content = prompt.to_content()
+        expected_content = [
+            Content(
+                parts=[
+                    Part.from_text("Which movie is better, {movie1} or {movie2}?"),
+                ],
+                role="user",
+            )
+        ]
+        assert unassembled_prompt_content[0].role == expected_content[0].role
+        for i in range(len(unassembled_prompt_content[0].parts)):
+            assert (
+                unassembled_prompt_content[0].parts[i].text
+                == expected_content[0].parts[i].text
+            )
+
+        # Check assembled prompt content
+        prompt.assemble()
+        assembled_prompt_content = prompt.to_content()
+        expected_content = [
+            Content(
+                parts=[
+                    Part.from_text("Which movie is better, "),
+                    Part.from_text("The Avengers"),
+                    Part.from_text(" or "),
+                    Part.from_text("Frozen"),
+                    Part.from_text("?"),
+                ],
+                role="user",
+            )
+        ]
+        assert assembled_prompt_content[0].role == expected_content[0].role
+        for i in range(len(assembled_prompt_content[0].parts)):
+            assert (
+                assembled_prompt_content[0].parts[i].text
+                == expected_content[0].parts[i].text
+            )
+
+    def test_inferred_prompt_assemble(self):
+        prompt = Prompt(
+            prompt_data="Which movie is better, {movie1} or {movie2}?",
+            variables={
+                "movie1": "The Avengers",
+            },
+        )
+
+        # Check partially assembled prompt content
+        prompt.assemble()
+        assembled1_prompt_content = prompt.to_content()
+        expected1_content = [
+            Content(
+                parts=[
+                    Part.from_text("Which movie is better, "),
+                    Part.from_text("The Avengers"),
+                    Part.from_text(" or "),
+                    Part.from_text("{movie2}"),
+                    Part.from_text("?"),
+                ],
+                role="user",
+            )
+        ]
+        assert assembled1_prompt_content[0].role == expected1_content[0].role
+        for i in range(len(assembled1_prompt_content[0].parts)):
+            assert (
+                assembled1_prompt_content[0].parts[i].text
+                == expected1_content[0].parts[i].text
+            )
+
+        # Check assembled prompt content
+        prompt.assemble(movie2="Frozen")
+        assembled2_prompt_content = prompt.to_content()
+        expected2_content = [
+            Content(
+                parts=[
+                    Part.from_text("Which movie is better, "),
+                    Part.from_text("The Avengers"),
+                    Part.from_text(" or "),
+                    Part.from_text("Frozen"),
+                    Part.from_text("?"),
+                ],
+                role="user",
+            )
+        ]
+        assert assembled2_prompt_content[0].role == expected2_content[0].role
+        for i in range(len(assembled2_prompt_content[0].parts)):
+            assert (
+                assembled2_prompt_content[0].parts[i].text
+                == expected2_content[0].parts[i].text
+            )
+
+        # Check assembled prompt content with override
+        prompt.assemble(movie1="Inception")
+        assembled3_prompt_content = prompt.to_content()
+        expected3_content = [
+            Content(
+                parts=[
+                    Part.from_text("Which movie is better, "),
+                    Part.from_text("Inception"),
+                    Part.from_text(" or "),
+                    Part.from_text("Frozen"),
+                    Part.from_text("?"),
+                ],
+                role="user",
+            )
+        ]
+        assert assembled3_prompt_content[0].role == expected3_content[0].role
+        for i in range(len(assembled3_prompt_content[0].parts)):
+            assert (
+                assembled3_prompt_content[0].parts[i].text
+                == expected3_content[0].parts[i].text
+            )
diff --git a/vertexai/generative_models/_prompt.py b/vertexai/generative_models/_prompt.py
new file mode 100644
index 00000000000..cb6af4ef895
--- /dev/null
+++ b/vertexai/generative_models/_prompt.py
@@ -0,0 +1,295 @@
+# -*- coding: utf-8 -*-
+
+# Copyright 2024 Google LLC
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+#
+
+from vertexai.generative_models import Content, Part
+
+import re
+from typing import (
+    Any,
+    Dict,
+    List,
+    Union,
+)
+
+VARIABLE_NAME_REGEX = r"(\{[^\W0-9]\w*\})"
+
+
+class ExpandablePartsVariable:
+    """A List[Parts] variable that can be expanded to multiple parts.
+
+    The name field should be the name of the variable, not wrapped with {}. This
+    class wraps a text Part with the name of the variable wrapped with {}.
+
+    Usage:
+        ```
+        prompt = Prompt(
+            prompt_data=[Content(
+                parts=[
+                    Part.from_text("What do these pictures have in common?"),
+                    ExpandablePartsVariable(name="pictures"),
+                ],
+                role="user"
+            )],
+            variables={"pictures": [
+                    Part.from_uri(uri="gs://.../img1.jpg", mime_type="image/jpeg"),
+                    Part.from_uri(uri="gs://.../img2.jpg", mime_type="image/jpeg"),
+            ]}
+        )
+        ```
+    """
+
+    def __init__(self, name: str):
+        """Initializes the ExpandablePartsVariable with a name.
+
+        Args:
+            name: The name of the variable, not wrapped with {}.
+        """
+        self.name = name
+        self._text_part = Part.from_text(f"{{{name}}}")
+        self._raw_part = self._text_part._raw_part
+
+    def to_part(self) -> Part:
+        """Returns the underlying text Part of the ExpandablePartsVariable.
+        The text is the name of the variable wrapped with {}.
+        """
+        return self._text_part
+
+
+class Prompt:
+    """A prompt which may be a template with variables.
+
+    The `Prompt` class allows users to define a template string with
+    variables represented in curly braces `{variable}`. The variable
+    name must be a valid Python variable name (no spaces, must start with a
+    letter). These placeholders can be replaced with specific values using the
+    `assemble` method, providing flexibility in generating dynamic prompts.
+
+    Usage:
+
+        ```
+        prompt_data =
+        prompt = Prompt(
+            prompt_data="Hello, {name}! Today is {day}. How are you?",
+            variables={"name": "Alice", "day": "Monday"}
+        )
+        prompt.assemble()
+        prompt.to_content()
+        ```
+    """
+
+    def __init__(
+        self,
+        prompt_data: Union[str, List[Content]],
+        variables: Dict[str, Union[str, List[Part]]],
+    ):
+        """Initializes the Prompt with a given prompt, and variables.
+
+        Args:
+            prompt: A string or List[Content] prompt which may be a template with variables or a prompt with no variables.
+            variables: A dictionary containing the variable names and values.
+        """
+        self._check_valid_prompt_data(prompt_data)
+        self.prompt_data = prompt_data
+        self._check_valid_variables(variables)
+        self.variables = self._format_variables(variables)
+        self._assembled_prompt = None
+
+    def set_prompt(self, prompt_data: Union[str, List[Content]]) -> None:
+        """Overwrites the existing prompt_data.
+
+        Args:
+            prompt_data: A string or List[Content] prompt.
+        """
+        self._check_valid_prompt_data(prompt_data)
+        self.prompt_data = prompt_data
+        self._assembled_prompt = None
+
+    def set_variables(self, variables: Dict[str, Union[str, List[Part]]]) -> None:
+        """Overwrites the existing variables dictionary.
+
+        Args:
+            variables: A dictionary containing the variable names and values.
+        """
+        self._check_valid_variables(variables)
+        self.variables = self._format_variables(variables)
+        self._assembled_prompt = None
+
+    def _format_variables(
+        self, variables: Dict[str, Union[str, List[Part]]]
+    ) -> Dict[str, List[Part]]:
+        """Formats the variables values to be List[Part]."""
+        for key in variables.keys():
+            if isinstance(variables[key], str):
+                variables[key] = [Part.from_text(variables[key])]
+        return variables
+
+    def _is_list_type(self, obj: Any, T: Any) -> bool:
+        return isinstance(obj, list) and all(isinstance(s, T) for s in obj)
+
+    def _check_valid_prompt_data(self, prompt_data: Any) -> None:
+        if not (
+            isinstance(prompt_data, str) or self._is_list_type(prompt_data, Content)
+        ):
+            raise ValueError(
+                "Prompt data must be a string or a list of Content objects."
+            )
+
+    def _check_valid_variables(self, variables: Any) -> None:
+        """Dict must be a Dict[str, Union[str, List[Part]]."""
+        valid = True
+        if isinstance(variables, dict):
+            for key, value in variables.items():
+                if not isinstance(key, str):
+                    valid = False
+                if not (isinstance(value, str) or self._is_list_type(value, Part)):
+                    valid = False
+        else:
+            valid = False
+        if not valid:
+            raise ValueError(
+                "Variable values must be a string or a list of Part objects."
+            )
+
+    def assemble(self, **kwargs) -> None:
+        """Replaces only the provided variables in the template with specific values.
+
+        Args:
+            **kwargs: Keyword arguments where keys are placeholder names and values
+              are the replacements.
+
+        Returns:
+            A new PromptTemplate instance with the updated template string.
+
+        Usage:
+            ```
+            prompt = Prompt(
+                prompt_data="Hello, {name}! Today is {day}. How are you?",
+                variables={"name": "Alice"}
+            )
+            prompt.assemble()
+            prompt.to_content()
+            # Returns "Hello, Alice! Today is {day}. How are you?" as List[Content]
+            prompt.assemble(day="Monday")
+            prompt.to_content()
+            # Returns "Hello, Alice! Today is Monday. How are you?" as List[Content]
+            prompt.assemble(name="Bob")
+            prompt.to_content()
+            # Returns "Hello, Bob! Today is Monday. How are you?" as List[Content]
+            ```
+        """
+        # Python Dict update will overwrite existing key values.
+        self.variables.update(kwargs)
+
+        # Convert the variables values to List[Part].
+        formatted_variables = self._format_variables(self.variables)
+        self._check_valid_variables(formatted_variables)
+
+        # Convert inferred prompt type (str) to static type (List[Content])
+        if isinstance(self.prompt_data, str):
+            # Step 1) Find and isolate variables as their own string.
+            prompt_data_str_split = re.split(VARIABLE_NAME_REGEX, self.prompt_data)
+
+            # Step 2) Replace each list element as a Part or ExpandablePartsVariable.
+            for i in range(len(prompt_data_str_split)):
+                if re.match(VARIABLE_NAME_REGEX, prompt_data_str_split[i]):
+                    prompt_data_str_split[i] = ExpandablePartsVariable(
+                        name=prompt_data_str_split[i][1:-1]
+                    )
+                else:
+                    prompt_data_str_split[i] = Part.from_text(prompt_data_str_split[i])
+
+            # Step 3) Wrap List[Part] as a single Content object.
+            list_content_prompt_data = [
+                Content(
+                    parts=prompt_data_str_split,
+                    role="user",
+                )
+            ]
+        elif self._is_list_type(self.prompt_data, Content):
+            list_content_prompt_data = self.prompt_data
+        else:
+            raise ValueError(
+                "Prompt data must be a string or a list of Content objects."
+            )
+
+        # Assemble the List[Content] prompt
+        for i in range(len(list_content_prompt_data)):
+            content = list_content_prompt_data[i]
+            expanded_parts = []
+            for part in content.parts:
+                # Infer part of format {variable_name} to be a ExpandablePartsVariable.
+                if (
+                    re.match(VARIABLE_NAME_REGEX, part.text)
+                    and part.text[1:-1] in formatted_variables
+                ):
+                    expanded_parts.extend(formatted_variables[part.text[1:-1]])
+                else:
+                    expanded_parts.append(part)
+
+            list_content_prompt_data[i] = Content(
+                parts=expanded_parts, role=content.role
+            )
+        self._assembled_prompt = list_content_prompt_data
+
+    def to_content(self) -> List[Content]:
+        """Returns the prompt data, assembled if prompt.assemble() was called.
+        Can be ingested into model.generate_content to make API calls.
+
+        Returns:
+            A List[Content] prompt.
+        Usage:
+            ```
+            prompt = Prompt(
+                prompt_data="Hello, {name}! Today is {day}. How are you?",
+                variables={"name": "Alice", "day": "Monday"}
+            )
+            prompt.assemble(day="Monday")
+            model.generate_content(
+                contents=prompt.to_content()
+            )
+            ```
+        """
+        if self._assembled_prompt:
+            # self._assembled_prompt must be List[Content]
+            return self._assembled_prompt
+        elif self._is_list_type(self.prompt_data, Content):
+            # If prompt_data is a List[Content], return it
+            return self.prompt_data
+        elif isinstance(self.prompt_data, str):
+            # If prompt_data is a string, wrap it with a Content object with singleton Part.
+            return [
+                Content(
+                    parts=[Part.from_text(self.prompt_data)],
+                    role="user",
+                )
+            ]
+        else:
+            raise ValueError(
+                "Prompt data must be a string or a list of Content objects."
+            )
+
+    def get_unassembled_prompt(self) -> Union[List[Content], str]:
+        """Returns the prompt data, without any variables replaced."""
+        return self.prompt_data
+
+    def __str__(self) -> str:
+        """Returns the prompt data, assembled if prompt.assemble() was called."""
+        return str(self._assembled_prompt) or self.prompt_data
+
+    def __repr__(self) -> str:
+        """Returns a string representation of the unassembled prompt."""
+        return f"Prompt(prompt_data='{self.prompt_data}', variables={self.variables})"