TooManyRequestsError when calling OpenAI API

[Eth3rnit3]

Problem Description

When running our application, we encounter a Faraday::TooManyRequestsError when calling the OpenAI API. This occurs when we try to read the content of a file named 'example.txt'.

Steps to Reproduce

1. Run the setup method to initialize the application.
2. Create a new instance of Langchain::LLM::OpenAI with the appropriate API key.
3. Create a new instance of Langchain::Thread.
4. Create a new instance of Langchain::Assistant with the appropriate parameters.
5. Add a message to the assistant asking to read the content of the 'example.txt' file.
6. Run the assistant.

Expected Outcome

The assistant should be able to read the content of the 'example.txt' file without encountering an error.

Actual Outcome

The application encounters a Faraday::TooManyRequestsError when calling the OpenAI API.

Additional Information

Here is the code:

def setup
  require 'bundler/setup'
  Bundler.require
  Dotenv.load
end

setup

llm = Langchain::LLM::OpenAI.new(api_key: ENV['OPENAI_API_KEY'])
thread = Langchain::Thread.new
assistant = Langchain::Assistant.new(
  llm:,
  thread:,
  instructions: 'You are a FileSystem Assistant that is able to read and write files. You can help with various file operations. To do this, you need clear and precise instructions. You will respond to requests in JSON format. How can you help today?',
  tools: [
    Langchain::Tool::FileSystem.new
  ]
)

assistant.add_message content: "I need to read the contents of a file named 'example.txt'."

assistant.run

Here is the error trace:

I, [2024-05-19T10:07:10.261834 #5338]  INFO -- : [Langchain.rb] [Langchain::Assistant]: Sending a call to Langchain::LLM::OpenAI
/Users/david/.rbenv/versions/3.2.2/lib/ruby/gems/3.2.0/gems/faraday-2.9.0/lib/faraday/response/raise_error.rb:34:in `on_complete': the server responded with status 429 (Faraday::TooManyRequestsError)
    from /Users/david/.rbenv/versions/3.2.2/lib/ruby/gems/3.2.0/gems/faraday-2.9.0/lib/faraday/middleware.rb:18:in `block in call'
    from /Users/david/.rbenv/versions/3.2.2/lib/ruby/gems/3.2.0/gems/faraday-2.9.0/lib/faraday/response.rb:42:in `on_complete'
    from /Users/david/.rbenv/versions/3.2.2/lib/ruby/gems/3.2.0/gems/faraday-2.9.0/lib/faraday/middleware.rb:17:in `call'
    from /Users/david/.rbenv/versions/3.2.2/lib/ruby/gems/3.2.0/gems/faraday-2.9.0/lib/faraday/rack_builder.rb:152:in `build_response'
    from /Users/david/.rbenv/versions/3.2.2/lib/ruby/gems/3.2.0/gems/faraday-2.9.0/lib/faraday/connection.rb:444:in `run_request'
    from /Users/david/.rbenv/versions/3.2.2/lib/ruby/gems/3.2.0/gems/faraday-2.9.0/lib/faraday/connection.rb:280:in `post'
    from /Users/david/.rbenv/versions/3.2.2/lib/ruby/gems/3.2.0/gems/ruby-openai-7.0.1/lib/openai/http.rb:22:in `json_post'
    from /Users/david/.rbenv/versions/3.2.2/lib/ruby/gems/3.2.0/gems/ruby-openai-7.0.1/lib/openai/client.rb:30:in `chat'
    from /Users/david/.rbenv/versions/3.2.2/lib/ruby/gems/3.2.0/gems/langchainrb-0.13.1/lib/langchain/llm/openai.rb:137:in `block in chat'
    from /Users/david/.rbenv/versions/3.2.2/lib/ruby/gems/3.2.0/gems/langchainrb-0.13.1/lib/langchain/llm/openai.rb:172:in `with_api_error_handling'
    from /Users/david/.rbenv/versions/3.2.2/lib/ruby/gems/3.2.0/gems/langchainrb-0.13.1/lib/langchain/llm/openai.rb:136:in `chat'
    from /Users/david/.rbenv/versions/3.2.2/lib/ruby/gems/3.2.0/gems/langchainrb-0.13.1/lib/langchain/assistants/assistant.rb:190:in `chat_with_llm'
    from /Users/david/.rbenv/versions/3.2.2/lib/ruby/gems/3.2.0/gems/langchainrb-0.13.1/lib/langchain/assistants/assistant.rb:93:in `run'
    from app.rb:22:in `<main>'

[Eth3rnit3]

Sorry it is due do my limit plan

[andreibondarev]

@Eth3rnit3 So you're verified and it works?

[Eth3rnit3]

Yes @andreibondarev it works fine 👍