WiseGPTEx

WiseGPTEx is an Elixir library that integrates with various language models, including OpenAI's GPT-3.5-turbo, Anthropic's Claude model, and Mistral's AI model. It provides intelligent question answering, reasoning capabilities, and AI-generated text completions across these platforms.

Installation

  1. Add wise_gpt_ex to your list of dependencies in mix.exs: 

    def deps do
[
 {:wise_gpt_ex, "~> 0.8.0"}
]
end

  2. Add the OpenAI, Anthropic, and Mistral API keys to your configuration file (e.g., config/config.exs): 

    config :wise_gpt_ex, :openai_api_key, "your_openai_api_key"
config :wise_gpt_ex, :anthropic_api_key, "your_anthropic_api_key"
config :wise_gpt_ex, :mistral_api_key, "your_mistral_api_key"
config :wise_gpt_ex, :groq_api_key, "your_groq_api_key"

Usage (only with OpenAI API)

To use WiseGPTEx, simply call the get_best_completion/2 or get_best_completion_with_resolver/2 function with a question and optional list of options:

{:ok, response} = WiseGPTEx.get_best_completion("What is the capital of France?")

You can also pass options to customize the API request:

opts = [model: "gpt-4", temperature: 0.4, num_completions: 4, timeout: 3_600_000]
{:ok, response} = WiseGPTEx.get_best_completion_with_resolver("What is the capital of France?", opts)

You can also make a raw call to the OpenAI API using the openai_completion/2 function:

messages = [
  %{"role" => "system", "content" => "You are High School Geography Teacher"},
  %{"role" => "user", "content" => "What was the capital of France in 15th century?"}
]
{ok, response} = WiseGPTEx.openai_completion(messages, [model: "gpt-4", temperature: 0.75, timeout: 3_600])

Note that the get_best_completion_with_resolver/2 function is similar to get_best_completion/2. This is because the difference between these two functions is in the method of how they select the best completion, not in their usage or the nature of their inputs or outputs. The get_best_completion_with_resolver/2 function will perform an additional API call to get a more accurate completion, which can be beneficial for complex or ambiguous queries. The openai_completion/2 function allows sending a custom conversation to the API without any additional prompting or setting the up the system role etc. This is useful when you want direct access to the model's system messages.

OpenAI API

For interactions with the OpenAI API, use the openai_completion/1 function:

{:ok, response} = WiseGPTEx.openai_completion("What is the capital of France?")
{:ok, "Paris"}

Anthropic API

For interactions with the Anthropic API, use the anthropic_completion/1 function:

{:ok, response} = WiseGPTEx.anthropic_completion("Why is the sky blue?")
{:ok, "The sky is blue because of the way sunlight interacts with Earth's atmosphere."}

Mistral API

For interactions with the Mistral API, use the mistral_completion/1 function:

{:ok, response} = WiseGPTEx.mistral_completion("What is the best French cheese?")
{:ok, "The best French cheese is..."}

Groq API

For interactions with the Groq API, use the groq_completion/2 function:

messages = [
  %{"role" => "system", "content" => "You are a helpful assistant."},
  %{"role" => "user", "content" => "What's the capital of France?"}
]
{:ok, response} = WiseGPTEx.groq_completion(messages)
{:ok,
 "The capital of France is Paris. Paris is a major European city and a global center for art, fashion, gastronomy, and culture. It is also the most populous city in France and the largest city in the European Union."}

Options

Options for the OpenAI API functions:

Options for the Anthropic API function:

Options for the Mistral API function:

Options for the Groq API function:

License

WiseGPTEx is licensed under the Apache License, Version 2.0. See the LICENSE file for more details.