stringableinference.py

from typing import Optional
import requests
import base64
import json


class StringableInference:
    """
    A client for interacting with the StringableInference API.
    
    Attributes:
        base_url (str): Base URL of the API.
        model (str): Model identifier to use for completions.
        session (requests.Session): Session object for making HTTP requests.
    """

    def __init__(self, model: str) -> None:
        """
        Initialize the client with a specific model.

        Args:
            model (str): The model identifier to use for API requests.
            
            (you can get available models using the get_models method)
        """
        self.base_url = "https://stringableinf.com/api"
        self.model = model
        self.session = requests.Session()

    def chat_completion(
        self,
        messages: list,
        stream: Optional[bool] = False,
        http_referer: Optional[str] = "https://bhattaraisusan.com.np",
    ) -> dict:
        """
        Send a chat completion request to the API.

        Args:
            messages (list): A list of message dictionaries to send.
            stream (Optional[bool]): Whether to enable streaming responses. Default is False.
            http_referer (Optional[str]): The HTTP referer header value. Default is a preset URL.

        Returns:
            dict: The JSON response from the API.

        Raises:
            Exception: If the HTTP request fails or the server returns an error.
        """
        try:
            response = self.session.post(
                self.base_url + "/v1/chat/completions",
                headers={
                    "Content-Type": "application/json",
                    "HTTP-Referer": http_referer,
                    "X-Title": "AI Assistant",
                },
                json={
                    "model": self.model,
                    "messages": messages,
                    "stream": stream,
                },
                stream=stream,
                timeout=15
            )

            response.raise_for_status()
            return json.loads(response.content)

        except Exception as e:
            print("Request failed:", e)
            raise

    def get_models(self) -> list:
        """
        Retrieve the list of available models from the API.

        Returns:
            list: A list of model backend IDs.

        Raises:
            Exception: If the HTTP request fails or the server returns an error.
        """
        try:
            response = self.session.get(self.base_url + "/models", timeout=15)
            response.raise_for_status()

            models_json = json.loads(response.content)
            models_list = [
                model["backend_id"] for model in models_json["models"]
            ]

            return models_list

        except Exception as e:
            print("Request failed:", e)
            raise

    def chat_completion_with_images(self) -> None:
        """
        Placeholder for chat completion with image support.

        Currently not supported by the provider. Intended to send messages
        containing images for processing.
        """
        # Not yet supported by the provider

        # with open("image.jpg", "rb") as f:
        #     image_data = base64.b64encode(f.read()).decode()
        # messages = [
        #     {
        #         "role": "user",
        #         "content": [
        #             {"type": "text", "text": "Describe this image"},
        #             {
        #                 "type": "image_url",
        #                 "image_url": {
        #                     "url": f"data:image/jpeg;base64,{image_data}"
        #                 },
        #             },
        #         ],
        #     }
        # ]
        pass