Moondream Ruby Client

A Ruby gem for interacting with the Moondream API, a powerful vision-language model for image understanding. This gem provides a simple interface for querying images, detecting objects, pointing to objects, and generating captions.

Add this line to your application's Gemfile:

gem 'moondream'

And then execute:

$ bundle install

Or install it yourself with:

$ gem install moondream

First, obtain an API key from Moondream. Then initialize the client:

require 'moondream'

client = Moondream::Client.new(api_key: "your_api_key_here")

Ask questions about an image using natural language:

response = client.query(
  "https://example.com/image.jpg",
  "What is in this image?"
)
puts response

Detect specific objects in an image and get their locations:

response = client.detect(
  "https://example.com/image.jpg",
  "person"
)
puts response

Get the coordinates of specific objects in an image:

response = client.point(
  "https://example.com/image.jpg",
  "the red car"
)
puts response

Generate descriptive captions for images:

# Normal length caption
response = client.caption(
  image_url: "https://example.com/image.jpg",
  length: "normal"
)
puts response

# Short caption
response = client.caption(
  image_url: "https://example.com/image.jpg",
  length: "short"
)
puts response

For real-time caption generation, you can use streaming mode:

client.caption(
  image_url: "https://example.com/image.jpg",
  length: "normal",
  stream: true
) do |response|
  response.read_body do |chunk|
    print chunk
  end
end

Creates a new Moondream client instance.

Parameters:

• api_key (String, required): Your Moondream API key

Query an image with a natural language prompt.

Parameters:

• image_url (String): URL of the image to analyze
• prompt (String): The question or prompt about the image

Returns: String response from the API

Detect objects in an image.

Parameters:

• image_url (String): URL of the image to analyze
• object (String): The object to detect

Returns: String response from the API with detection results

Get coordinates pointing to specific objects in an image.

Parameters:

• image_url (String): URL of the image to analyze
• object (String): The object to point to

Returns: String response from the API with coordinate information

Generate a caption for an image.

Parameters:

• image_url (String, required): URL of the image to caption
• length (String, optional): Caption length - "normal" or "short" (default: "normal")
• stream (Boolean, optional): Enable streaming mode (default: false)

Returns: String response from the API, or yields response chunks if streaming with a block

The gem raises Moondream::Error for API-related errors. Wrap your calls in error handling:

begin
  response = client.query(image_url, prompt)
  puts response
rescue Moondream::Error => e
  puts "Error: #{e.message}"
end

After checking out the repo, run bin/setup to install dependencies. Then, run rake spec to run the tests. You can also run bin/console for an interactive prompt that will allow you to experiment.

To install this gem onto your local machine, run bundle exec rake install. To release a new version, update the version number in version.rb, and then run bundle exec rake release, which will create a git tag for the version, push git commits and the created tag, and push the .gem file to rubygems.org.

Bug reports and pull requests are welcome on GitHub at https://github.com/yourusername/moondream-ruby.

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