Userapi.ai is an awesome tool that brings great value to products. This gem was created to facilitate Ruby and also Rails applications to use all services of Userapi.ai:
- Midjourney API
Add the following to your Gemfile:
gem "userapi-ai"Or install with:
gem install userapi-aiand require with:
require "userapi-ai"Run bundle install.
That's it. You are now ready to go!
Make sure you have a
- A Userapi.ai account at Userapi.ai
- Discord Account or create one at Discord
- A Midjourney account or sign up at Midjourney
- A Midjourney Paid plan to use this gem or sign up at Midjourney Plans
After signing in, you can get your Discord token by checking out any message request you send in the browser.
You can do this by opening the developer tools in your browser and going to the network tab.
Then send a message in any channel and look for the request to messages in the network tab.
Check out the Request Headers and look for the authorization header. Copy the value of the header and use it as your Discord User Tokens. Read more here.
Once you have your Userapi.ai Token, you can configure the gem with:
Open up the initializer file config/initializers/userapi.rb and add your API Key as the following:
UserapiAi.configure do |config|
config.access_token = ENV["USERAPI_USER_TOKEN"]
endImagine is a Midjourney command that allows you to use AI to generate images. You can use it to generate images of people, animals, objects, and anything else you can imagine.
client = UserapiAi::Client.new(service: :midjourney)
result = client.imagine(prompt: "nice girl")
# or
result = client.imagine(prompt: "nice girl",
webhook_url: "https://example.com/imagine/webhook-url",
webhook_type: "progress", # or result
account_hash: "a7d44910-88a6-4673-acc8-d60ffc3479a6",
is_disable_prefilter: false)
# then
client.status(result) client = UserapiAi::Client.new(service: :midjourney)
result = client.status(hash: "xxx-xxx")
result = client.variation(hash: "xxx-xxx", choice: 1)
result = client.upsample(hash: "xxx-xxx", choice: "v6_2x_subtle")
result = client.upscale(hash: "xxx-xxx",
choice: 1,
webhook_url: "https://example.com/upscale/webhook-url",
webhook_type: "result")See more here.
You can split the quad-layout output from Midjourney into 4 separate images.
Add the gem to your Gemfile.
gem "mini_magick"Now you can do some magick.
response = client.status result
if response["status"] == "done"
local_paths = UserapiAi::Tool.crop result: response["result"]
# or UserapiAi::Tool.crop url: "https://example.com/img.png"
local_paths.each do |path|
f = File.open(path)
...
end
end=> local_paths
["/var/folders/yb/bjdvl6mn3cx2l7nd08726k500000gn/T/mini_magick20240324-73984-31zmkb-0.png",
"/var/folders/yb/bjdvl6mn3cx2l7nd08726k500000gn/T/mini_magick20240324-73984-31zmkb-1.png",
"/var/folders/yb/bjdvl6mn3cx2l7nd08726k500000gn/T/mini_magick20240324-73984-31zmkb-2.png",
"/var/folders/yb/bjdvl6mn3cx2l7nd08726k500000gn/T/mini_magick20240324-73984-31zmkb-3.png"]After checking out the repo, run bin/setup to install dependencies. Then, run bundle exec rspec to run the tests.
You can also run bin/console for an interactive prompt that will allow you to experiment.
Bug reports and pull requests are welcome on GitHub at Open an Issue.
The gem is available as open source under the terms of the MIT License. Feel free to use it and contribute.