This guide explains how to seamlessly integrate the powerful Gemini AI API into your Ruby projects. Utilize Gemini's cutting-edge language capabilities for generating text, translating languages, and more.

• Installation
  • Bundler
  • Gem Install
• Credentials
  • Option 1: API Key (Generative Language API)
  • Option 2: Service Account Credentials File (Vertex AI API)
  • Option 3: Application Default Credentials (Vertex AI API)
• Usage
  • Quickstart
  • With Config
  • Verbose Logging
• Methods
  • stream_generate_content
  • generate_content
• Development
• Compatibility
• License
• Resources and References
• Additional Notes

Add this line to your application's Gemfile:

gem "ruby-gemini-ai"

And then execute:

$ bundle install

Or install with:

$ gem install ruby-gemini-ai

and require with:

require "gemini-ai"

• Option 1: API Key (Generative Language API)
• Option 2: Service Account Credentials File (Vertex AI API)
• Option 3: Application Default Credentials (Vertex AI API)

Obtain an API Key from your Google Cloud project: Google Cloud through the Google Cloud Console: https://console.cloud.google.com/apis/credentials.

Enable the Generative Language API service in your Google Cloud Console. which can be done here.

Alternatively, you can generate an API Key through Google AI Studio here, which will automatically create a project for you.

For the Vertex AI API, create a Google Cloud Project and a Service Account. Enable the [Vertex AI] (https://cloud.google.com/vertex-ai) API for your project here.

Generate credentials for your Service Account here and download a JSON file named google-credentials.json.

{
  "type": "service_account",
  "project_id": "YOUR_PROJECT_ID",
  "private_key_id": "a00...",
  "private_key": "-----BEGIN PRIVATE KEY-----\n...\n-----END PRIVATE KEY-----\n",
  "client_email": "PROJECT_ID@PROJECT_ID.iam.gserviceaccount.com",
  "client_id": "000...",
  "auth_uri": "https://accounts.google.com/o/oauth2/auth",
  "token_uri": "https://oauth2.googleapis.com/token",
  "auth_provider_x509_cert_url": "https://www.googleapis.com/oauth2/v1/certs",
  "client_x509_cert_url": "https://www.googleapis.com/robot/v1/metadata/x509/..."
}

Ensure the necessary policies (roles/aiplatform.user and possibly roles/ml.admin) are in place use the Vertex AI API.

You can add them by navigating to the IAM Console and clicking on the "Edit principal" (✏️ pencil icon) next to your Service Account.

Alternatively, you can add them through the gcloud CLI as follows:

gcloud projects add-iam-policy-binding PROJECT_ID \
  --member='serviceAccount:PROJECT_ID@PROJECT_ID.iam.gserviceaccount.com' \
  --role='roles/aiplatform.user'

Some people reported having trouble accessing the API, and adding the role roles/ml.admin fixed it:

gcloud projects add-iam-policy-binding PROJECT_ID \
  --member='serviceAccount:PROJECT_ID@PROJECT_ID.iam.gserviceaccount.com' \
  --role='roles/ml.admin'

If you are not using a Service Account:

gcloud projects add-iam-policy-binding PROJECT_ID \
  --member='user:[email protected]' \
  --role='roles/aiplatform.user'

gcloud projects add-iam-policy-binding PROJECT_ID \
  --member='user:[email protected]' \
  --role='roles/ml.admin'

Similar to Option 2, but you don't need to download a google-credentials.json. These automatically find credentials based on your environment. Application Default Credentials.

Generate them using the gcloud CLI before local development. gcloud CLI:

gcloud auth application-default login 

For more details about alternative methods and different environments, check the official documentation: Set up Application Default Credentials

For a quick test you can pass your token directly to a new client:

client = GeminiAi::Client.new(api_key: "gemini_api_key")

We can configure Gemini with Ruby using three options.

Option 1, API KEY

For a more robust setup, you can configure the gem with your API keys, for example in an gemini.rb initializer file. Never hardcode secrets into your codebase - instead use something like dotenv to pass the keys safely into your environments.

GeminiAi.configure do |config|
  config.api_key = ENV.fetch("GEMINI_API_KEY")
  config.service = ENV.fetch("GEMINI_API_SERVICE")
end

Option 2, Service Account

For the Service Account, provide a google-credentials.json file and a REGION:

GeminiAi.configure do |config|
  config.service = 'vertex-ai-api'
  config.region = 'us-east4'
  config.file_path = 'google-credentials.json'
end

Option 3, Default Credentials

For Application Default Credentials, omit both the api_key and the file_path:

GeminiAi.configure do |config|
  config.region = 'us-east4'
  config.service = 'vertex-ai-api'
end

Then you can create a client like this:

client = GeminiAi::Client.new

• Streams generated text in real-time.
• contents (hash): User input and role information.
• model (string): Optional model name (e.g., gemini-pro).
• Returns an array of candidates objects with generated text and safety ratings.

client = GeminiAi::Client.new
# Assuming you configured with your API key or credentials

contents = {
  contents: {
    role: 'user',
    parts: {
      text: 'Write a poem about the ocean.'
    }
  }
}

stream = client.stream_generate_content(contents, model: 'gemini-pro')

In this case, the result will be an array with all the received events:

[{ 'candidates' =>
   [{ 'content' => {
        'role' => 'model',
        'parts' => [{ 'text' => 'exmaple poem content.......' }]
      },
      'finishReason' => 'STOP',
      'safetyRatings' =>
      [{ 'category' => 'HARM_CATEGORY_HARASSMENT', 'probability' => 'NEGLIGIBLE' },
       { 'category' => 'HARM_CATEGORY_HATE_SPEECH', 'probability' => 'NEGLIGIBLE' },
       { 'category' => 'HARM_CATEGORY_SEXUALLY_EXPLICIT', 'probability' => 'NEGLIGIBLE' },
       { 'category' => 'HARM_CATEGORY_DANGEROUS_CONTENT', 'probability' => 'NEGLIGIBLE' }] }],
   'usageMetadata' => {
     'promptTokenCount' => 2,
     'candidatesTokenCount' => 8,
     'totalTokenCount' => 10
   } }]

client = GeminiAi::Client.new
# Assuming you configured with your API key or credentials

contents = {
  contents: {
    role: 'user',
    parts: {
      text: 'Write a poem about the ocean.'
    }
  }
}

client.stream_generate_content(contents, model: 'gemini-pro', stream: true) do |part_text, event, parsed, raw|
  puts text
end

OR

client = GeminiAi::Client.new
# Assuming you configured with your API key or credentials

contents = {
  contents: {
    role: 'user',
    parts: {
      text: 'Write a poem about the ocean.'
    }
  }
}

# Assuming you have a block or procedure (proc) defined
stream_proc = Proc.new do |part_text, event, parsed, raw|
  puts part_text 
end

client.stream_generate_content(contents, model: 'gemini-pro', stream: true, &stream_proc)

In this case, the result will be an array with all the received events:

'exmaple poem content.......'

• Generates text in a single request.
• contents (hash): User input and role information.
• model (string): Optional model name (e.g., gemini-pro).
• Returns a hash with generated text, safety ratings, and prompt feedback

result = client.generate_content(
  { contents: { role: 'user', parts: { text: 'hi!' } } }, model: 'gemini-pro'
)

client = GeminiAi::Client.new
# Assuming you configured with your API key or credentials

contents = {
  contents: {
    role: 'user',
    parts: {
      text: 'Write a poem about the ocean.'
    }
  }
}

stream = client.generate_content(contents, model: 'gemini-pro')

Result:

{ 'candidates' =>
  [{ 'content' => { 'parts' => [{ 'text' => 'exampled poem.......' }], 'role' => 'model' },
     'finishReason' => 'STOP',
     'index' => 0,
     'safetyRatings' =>
     [{ 'category' => 'HARM_CATEGORY_SEXUALLY_EXPLICIT', 'probability' => 'NEGLIGIBLE' },
      { 'category' => 'HARM_CATEGORY_HATE_SPEECH', 'probability' => 'NEGLIGIBLE' },
      { 'category' => 'HARM_CATEGORY_HARASSMENT', 'probability' => 'NEGLIGIBLE' },
      { 'category' => 'HARM_CATEGORY_DANGEROUS_CONTENT', 'probability' => 'NEGLIGIBLE' }] }],
  'promptFeedback' =>
  { 'safetyRatings' =>
    [{ 'category' => 'HARM_CATEGORY_SEXUALLY_EXPLICIT', 'probability' => 'NEGLIGIBLE' },
     { 'category' => 'HARM_CATEGORY_HATE_SPEECH', 'probability' => 'NEGLIGIBLE' },
     { 'category' => 'HARM_CATEGORY_HARASSMENT', 'probability' => 'NEGLIGIBLE' },
     { 'category' => 'HARM_CATEGORY_DANGEROUS_CONTENT', 'probability' => 'NEGLIGIBLE' }] } }

You can pass Faraday middleware to the client in a block, eg. to enable verbose logging with Ruby's Logger:

  client = GeminiAi::Client.new do |f|
    f.response :logger, Logger.new($stdout), bodies: true
  end

1. Clone the repository.
2. Run bin/setup to install dependencies.
3. Use bin/console for interactive exploration.
4. Run bundle exec rake install to install the gem locally.

ruby-gemini-ai gem is compatible with Ruby versions 2.6.7 and higher.

The gem is available as open source under the terms of the MIT License.

Explore the following curated list of resources and references to enhance your understanding throughout the learning process:

• Google AI for Developers: Stay updated on the latest developments and resources in the field of Artificial Intelligence by visiting Google AI for Developers.
• Get started with the Gemini API: Initiate your journey into the Gemini API with comprehensive guides and documentation provided by Google.
• Getting Started with the Vertex AI Gemini API with cURL: Explore hands-on examples and tutorials using cURL to kickstart your experience with the Vertex AI Gemini API.
• Gemini API Documentation: Refer to the official Gemini API documentation for detailed information on model references and implementation guidelines.
• Vertex AI API Documentation: Dive into the Vertex AI API documentation to gain a comprehensive understanding of Vertex AI services.
• REST Documentation: Explore the RESTful API documentation for Vertex AI to facilitate seamless integration with your applications.
• Google DeepMind Gemini: Gain insights into Google DeepMind's Gemini technology, a cutting-edge advancement in the field of Artificial Intelligence.
• Stream responses from Generative AI models: Learn how to effectively stream responses from Generative AI models by consulting this specific guide within the Vertex AI documentation.
• Function calling: Understand the intricacies of function calling in the context of Generative AI models with this guide from the Vertex AI documentation.

These resources collectively provide a comprehensive foundation for your exploration of the Gemini API and Vertex AI services.

• As of now, only generate_content is supported with the vertex-ai-api service.
• For detailed API documentation and advanced usage, refer to the official Gemini AI documentation Consider adding examples and error handling for a more user-friendly experience.