Obiguard provides a robust and secure gateway to facilitate the integration of various Large Language Models (LLMs) into your applications, including Google Gemini APIs.

With Obiguard, you can take advantage of features like fast AI gateway access, observability, prompt management, and more, all while ensuring the secure management of your LLM API keys through a virtual key system.

Provider Slug. google

​
Obiguard SDK Integration with Google Gemini Models

Obiguard provides a consistent API to interact with models from various providers. To integrate Google Gemini with Obiguard:

​
1. Install the Obiguard SDK

Add the Obiguard SDK to your application to interact with Google Gemini’s API through Obiguard’s gateway.

pip install obiguard

​
2. Initialize Obiguard with the Virtual Key

To use Gemini with Obiguard, get your API key from here, then add it to Obiguard to create the virtual key.

from obiguard import Obiguard

client = Obiguard(
  obiguard_api_key="vk-obg***",  # Your Obiguard virtual key
)

​
3. Invoke Chat Completions with Google Gemini

Use the Obiguard instance to send requests to Google Gemini. You can also override the virtual key directly in the API call if needed.

completion = client.chat.completions.create(
    messages= [
        { "role": 'system', "content": 'You are not a helpful assistant' },
        { "role": 'user', "content": 'Say this is a test' }
    ],
    model= 'gemini-1.5-pro'
)

print(completion)

Obiguard supports the system_instructions parameter for Google Gemini 1.5 - allowing you to control the behavior and output of your Gemini-powered applications with ease.

Simply include your Gemini system prompt as part of the {"role":"system"} message within the messages array of your request body. Obiguard Gateway will automatically transform your message to ensure seamless compatibility with the Google Gemini API.

​
Function Calling

Obiguard supports function calling mode on Google’s Gemini Models.

​
Document, Video, Audio Processing with Gemini

Gemini supports attaching mp4, pdf, jpg, mp3, wav, etc. file types to your messages.

Gemini Docs:

Using Obiguard, here’s how you can send these media files:

Python
completion = client.chat.completions.create(
  messages=[
    {
      "role": "system",
      "content": "You are a helpful assistant"
    },
    {
      "role": "user",
      "content": [
        {
          "type": "image_url",
          "image_url": {
            "url": "gs://cloud-samples-data/generative-ai/image/scones.jpg"
          }
        },
        {
          "type": "text",
          "text": "Describe the image"
        }
      ]
    }
  ],
  model='gemini-1.5-pro',
  max_tokens=200
)
print(completion)

This same message format also works for all other media types — just send your media file in the url field, like "url": "gs://cloud-samples-data/video/animals.mp4".

Your URL should have the file extension, this is used for inferring MIME_TYPE which is a required parameter for prompting Gemini models with files.

​
Sending base64 Image

Here, you can send the base64 image data along with the url field too:

"url": "data:image/png;base64,UklGRkacAABXRUJQVlA4IDqcAAC....."

Vertex AI supports grounding with Google Search. This is a feature that allows you to ground your LLM responses with real-time search results. Grounding is invoked by passing the google_search tool (for newer models like gemini-2.0-flash-001), and google_search_retrieval (for older models like gemini-1.5-flash) in the tools array.

"tools": [
    {
        "type": "function",
        "function": {
            "name": "google_search" // or google_search_retrieval for older models
        }
    }]

If you mix regular tools with grounding tools, vertex might throw an error saying only one tool can be used at a time.

​
Extended Thinking (Reasoning Models) (Beta)

The assistants thinking response is returned in the response_chunk.choices[0].delta.content_blocks array, not the response.choices[0].message.content string.

Models like gemini-2.5-flash-preview-04-17 gemini-2.5-flash-preview-04-17 support extended thinking. This is similar to openai thinking, but you get the model’s reasoning as it processes the request as well.

Note that you will have to set strict_open_ai_compliance=False in the headers to use this feature.

​
Single turn conversation

Python
from obiguard import Obiguard

client = Obiguard(
  obiguard_api_key="vk-obg***",  # Your Obiguard virtual key
  strict_open_ai_compliance=False
)

# Create the request
response = client.chat.completions.create(
  model="gemini-2.5-flash-preview-04-17",
  max_tokens=3000,
  thinking={
    "type": "enabled",
    "budget_tokens": 2030
  },
  stream=True,
  messages=[
    {
      "role": "user",
      "content": [
        {
          "type": "text",
          "text": "when does the flight from new york to bengaluru land tomorrow, what time, what is its flight number, and what is its baggage belt?"
        }
      ]
    }
  ]
)

print(response)
# in case of streaming responses you'd have to parse the response_chunk.choices[0].delta.content_blocks array
# response = client.chat.completions.create(
#   ...same config as above but with stream: true
# )
# for chunk in response:
#     if chunk.choices[0].delta:
#         content_blocks = chunk.choices[0].delta.get("content_blocks")
#         if content_blocks is not None:
#             for content_block in content_blocks:
#                 print(content_block)

To disable thinking for gemini models like gemini-2.5-flash-preview-04-17, you are required to explicitly set budget_tokens to 0.

"thinking": {
    "type": "enabled",
    "budget_tokens": 0
}

Gemini grounding mode may not work via Obiguard SDK. Contact [email protected] for assistance.

​
Next Steps

The complete list of features supported in the SDK are available on the link below.

SDK