requestyai 0.3.4

Deliver AI products in days, not months, using requesty.ai

RequestyAI Python API library

Deliver AI products in days, not months, using requesty.ai

Step up your AI game with a simple integration:

• Real-time insights into your AI flows: See how your clients use your LLMs
• Hassle free auditing and logging: Don't stress about storage and infra
• Powerful analytics: Discover trends using your Requesty's insight-explorer

Installation

pip install requestyai

AInsights

A client library to effortlessly integrate powerful insights into your AI flows.

Create a client with minimal configuration and capture and request/response pair for future analysis.

Notes

Thread-safety

The insights client is thread-safe. You can safely use a single client from multiple threads.

Asynchronous

The insights client is asynchronous. capture()-ing requests/responses is a non-blocking operation, and it will not interrupt the flow of your application.

Usage pattern #1: Create client instance

If you prefer creating and using client instance, just add an AInsights instance next to your OpenAI one, and capture every interaction by adding a single call.

It doesn't matter if you're using simple free-text outputs, JSON outputs or tools, the insights client will capture everything.

Check out this working sample

from openai import OpenAI
from requestyai import AInsights


class OpenAIWrapper:
    def __init__(self, openai_api_key, requesty_api_key):
        self.model = OpenAI(api_key=openai_api_key)
        self.args = {
            model="gpt-4o",
            temperature=0.7,
            max_tokens=150
        }

        self.insights = AInsights.new_client(api_key=requesty_api_key)

    def chat(self, user_id: str, user_input: str):
        messages = [
            { "role": "system", "content": "You are a helpful assistant." },
            { "role": "user", "content": user_input }
        ]

        class Response(BaseModel):
            answer: str

        response = self.model.beta.chat.completions.parse(messages=messages,
                                                          response_format=Response,
                                                          **self.args)

        self.insights.capture(messages=messages,
                              response=response,
                              args=self.args,
                              user_id=user_id)

        return Response.model_validate_json(response.choices[0].message.content)


if __name__ == "__main__":
    requesty_api_key = os.environ["REQUESTY_API_KEY"]
    openai_api_key = os.environ["OPENAI_API_KEY"]

    wrapper = OpenAIWrapper(openai_api_key, requesty_api_key)

    # Dummy loggin to capture the user's ID
    user_id = input("User ID: ")

    while True:
        user_input = input("Ask any question?")
        response = wrapper.chat(user_id, user_input)
        print(response.answer)

Usage pattern #2: Use a global instance

You like it simple. You just want to call OpenAI from anywhere in your code, and the insights should be no different.

Just create a simple file (ainsights_instance.py is a reasonable name) in your project, and import it everywhere. The client is thread-safe.

Check out this working sample

import os
from requestyai import AInsights


requesty_key = os.environ["REQUESTY_KEY"]
ainsights = AInsights.new_client(api_key=requesty_key)

Then, calling OpenAI and capturing is as easy as:

It doesn't matter if you're using simple free-text outputs, JSON outputs or tools, the insights client will capture everything.

from openai import OpenAI
from ainsights_instance import ainsights


if __name__ == '__main__':
    openai.api_key = os.environ["OPENAI_API_KEY"]

    messages = [{ "role": "user", "content": user_input }]
    args = {
        model="gpt-4o-mini",
        temperature=0.7,
        max_tokens=150
    }
    response = openai.chat.completions.create(messages=messages, **args)

    ainsights.capture(messages=messages, response=response, args=args)

    print(response.choices[0].message.content)

User tracking

If you want your insights to be tied to a specific user, you can specify the user_id argument when calling capture(...).

See "Usage pattern #1" above for specific details.

Sample applications

Check out the samples directory for working examples you can try out in no time.