1  ---
  2  title: "VertexAIGeminiChatGenerator"
  3  id: vertexaigeminichatgenerator
  4  slug: "/vertexaigeminichatgenerator"
  5  description: "`VertexAIGeminiChatGenerator` enables chat completion using Google Gemini models."
  6  ---
  7  
  8  # VertexAIGeminiChatGenerator
  9  
 10  `VertexAIGeminiChatGenerator` enables chat completion using Google Gemini models.
 11  
 12  :::warning[Deprecation Notice]
 13  
 14  This integration uses the deprecated google-generativeai SDK, which will lose support after August 2025.
 15  
 16  We recommend switching to the new [GoogleGenAIChatGenerator](googlegenaichatgenerator.mdx) integration instead.
 17  :::
 18  
 19  <div className="key-value-table">
 20  
 21  |  |  |
 22  | --- | --- |
 23  | **Most common position in a pipeline** | After a [ChatPromptBuilder](../builders/chatpromptbuilder.mdx)                                                 |
 24  | **Mandatory run variables**            | `messages`: A list of [`ChatMessage`](../../concepts/data-classes/chatmessage.mdx)  objects representing the chat |
 25  | **Output variables**                   | `replies`: A list of alternative replies of the model to the input chat                              |
 26  | **API reference**                      | [Google Vertex](/reference/integrations-google-vertex)                                                      |
 27  | **GitHub link**                        | https://github.com/deepset-ai/haystack-core-integrations/tree/main/integrations/google_vertex      |
 28  
 29  </div>
 30  
 31  `VertexAIGeminiGenerator` supports `gemini-1.5-pro` and `gemini-1.5-flash`/  `gemini-2.0-flash` models. Note that [Google recommends upgrading](https://cloud.google.com/vertex-ai/generative-ai/docs/learn/model-versions) from `gemini-1.5-pro` to `gemini-2.0-flash`.
 32  
 33  For available models, see https://cloud.google.com/vertex-ai/generative-ai/docs/learn/models.
 34  
 35  :::info
 36  To explore the full capabilities of Gemini check out this [article](https://haystack.deepset.ai/blog/gemini-models-with-google-vertex-for-haystack) and the related [🧑‍🍳 Cookbook](https://colab.research.google.com/github/deepset-ai/haystack-cookbook/blob/main/notebooks/vertexai-gemini-examples.ipynb).
 37  :::
 38  
 39  ### Parameters Overview
 40  
 41  `VertexAIGeminiChatGenerator` uses Google Cloud Application Default Credentials (ADCs) for authentication. For more information on how to set up ADCs, see the [official documentation](https://cloud.google.com/docs/authentication/provide-credentials-adc).
 42  
 43  Keep in mind that it’s essential to use an account that has access to a project authorized to use Google Vertex AI endpoints.
 44  
 45  You can find your project ID in the [GCP resource manager](https://console.cloud.google.com/cloud-resource-manager) or locally by running `gcloud projects list` in your terminal. For more info on the gcloud CLI, see its [official documentation](https://cloud.google.com/cli).
 46  
 47  ### Streaming
 48  
 49  This Generator supports [streaming](guides-to-generators/choosing-the-right-generator.mdx#streaming-support) the tokens from the LLM directly in output. To do so, pass a function to the `streaming_callback` init parameter.
 50  
 51  ## Usage
 52  
 53  You need to install the `google-vertex-haystack` package to use the  `VertexAIGeminiChatGenerator`:
 54  
 55  ```shell
 56  pip install google-vertex-haystack
 57  ```
 58  
 59  ### On its own
 60  
 61  Basic usage:
 62  
 63  ```python
 64  from haystack.dataclasses import ChatMessage
 65  from haystack_integrations.components.generators.google_vertex import VertexAIGeminiChatGenerator
 66  
 67  gemini_chat = VertexAIGeminiChatGenerator()
 68  
 69  messages = [ChatMessage.from_user("Tell me the name of a movie")]
 70  res = gemini_chat.run(messages)
 71  
 72  print(res["replies"][0].text)
 73  >>> The Shawshank Redemption
 74  
 75  messages += [res["replies"][0], ChatMessage.from_user("Who's the main actor?")]
 76  res = gemini_chat.run(messages)
 77  
 78  print(res["replies"][0].text)
 79  >>> Tim Robbins
 80  ```
 81  
 82  When chatting with Gemini Pro, you can also easily use function calls. First, define the function locally and convert into a [Tool](../../tools/tool.mdx):
 83  
 84  ```python
 85  from typing import Annotated
 86  from haystack.tools import create_tool_from_function
 87  
 88  
 89  ## example function to get the current weather
 90  def get_current_weather(
 91      location: Annotated[
 92          str,
 93          "The city for which to get the weather, e.g. 'San Francisco'",
 94      ] = "Munich",
 95      unit: Annotated[str, "The unit for the temperature, e.g. 'celsius'"] = "celsius",
 96  ) -> str:
 97      return f"The weather in {location} is sunny. The temperature is 20 {unit}."
 98  
 99  
100  tool = create_tool_from_function(get_current_weather)
101  ```
102  
103  Create a new instance of `VertexAIGeminiChatGenerator` to set the tools and a [ToolInvoker](../tools/toolinvoker.mdx) to invoke the tools.:
104  
105  ```python
106  from haystack_integrations.components.generators.google_vertex import (
107      VertexAIGeminiChatGenerator,
108  )
109  from haystack.components.tools import ToolInvoker
110  
111  gemini_chat = VertexAIGeminiChatGenerator(model="gemini-2.0-flash-exp", tools=[tool])
112  
113  tool_invoker = ToolInvoker(tools=[tool])
114  ```
115  
116  And then ask our question:
117  
118  ```python
119  from haystack.dataclasses import ChatMessage
120  
121  messages = [ChatMessage.from_user("What is the temperature in celsius in Berlin?")]
122  res = gemini_chat.run(messages=messages)
123  
124  print(res["replies"][0].tool_calls)
125  >>> [ToolCall(tool_name='get_current_weather',
126  >>>           arguments={'unit': 'celsius', 'location': 'Berlin'}, id=None)]
127  
128  tool_messages = tool_invoker.run(messages=replies)["tool_messages"]
129  messages = user_message + replies + tool_messages
130  
131  messages += res["replies"][0] + [ChatMessage.from_function(content=weather, name="get_current_weather")]
132  
133  final_replies = gemini_chat.run(messages=messages)["replies"]
134  print(final_replies[0].text)
135  >>> The temperature in Berlin is 20 degrees Celsius.
136  ```
137  
138  ### In a pipeline
139  
140  ```python
141  from haystack.components.builders import ChatPromptBuilder
142  from haystack.dataclasses import ChatMessage
143  from haystack import Pipeline
144  from haystack_integrations.components.generators.google_vertex import VertexAIGeminiChatGenerator
145  
146  ## no parameter init, we don't use any runtime template variables
147  prompt_builder = ChatPromptBuilder()
148  gemini_chat = VertexAIGeminiChatGenerator()
149  
150  pipe = Pipeline()
151  pipe.add_component("prompt_builder", prompt_builder)
152  pipe.add_component("gemini", gemini)
153  pipe.connect("prompt_builder.prompt", "gemini.messages")
154  
155  location = "Rome"
156  messages = [ChatMessage.from_user("Tell me briefly about {{location}} history")]
157  res = pipe.run(data={"prompt_builder": {"template_variables":{"location": location}, "template": messages}})
158  
159  print(res)
160  
161  >>> - **753 B.C.:** Traditional date of the founding of Rome by Romulus and Remus.
162  >>> - **509 B.C.:** Establishment of the Roman Republic, replacing the Etruscan monarchy.
163  >>> - **492-264 B.C.:** Series of wars against neighboring tribes, resulting in the expansion of the Roman Republic's territory.
164  >>> - **264-146 B.C.:** Three Punic Wars against Carthage, resulting in the destruction of Carthage and the Roman Republic becoming the dominant power in the Mediterranean.
165  >>> - **133-73 B.C.:** Series of civil wars and slave revolts, leading to the rise of Julius Caesar.
166  >>> - **49 B.C.:** Julius Caesar crosses the Rubicon River, starting the Roman Civil War.
167  >>> - **44 B.C.:** Julius Caesar is assassinated, leading to the Second Triumvirate of Octavian, Mark Antony, and Lepidus.
168  >>> - **31 B.C.:** Battle of Actium, where Octavian defeats Mark Antony and Cleopatra, becoming the sole ruler of Rome.
169  >>> - **27 B.C.:** The Roman Republic is transformed into the Roman Empire, with Octavian becoming the first Roman emperor, known as Augustus.
170  >>> - **1st century A.D.:** The Roman Empire reaches its greatest extent, stretching from Britain to Egypt.
171  >>> - **3rd century A.D.:** The Roman Empire begins to decline, facing internal instability, invasions by Germanic tribes, and the rise of Christianity.
172  >>> - **476 A.D.:** The last Western Roman emperor, Romulus Augustulus, is overthrown by the Germanic leader Odoacer, marking the end of the Roman Empire in the West.
173  ```
174  
175  ## Additional References
176  
177  🧑‍🍳 Cookbook: [Function Calling and Multimodal QA with Gemini](https://haystack.deepset.ai/cookbook/vertexai-gemini-examples)