models_api.py

# import requests
# import json
# import os
# from dotenv import load_dotenv
# load_dotenv()

# def get_answer(model_name, context, question):    
#     llm_key = os.getenv("llm_key")
#     url = os.getenv("main_url")
#     # Construct the prompt for the model
#     prompt = f"You are a Question Answering Model. Can you help me answer the question: {question} from the context: {context}? Just return the answer only. The document may contain some Arabic text; please translate that to English if needed."

#     # Prepare payload for API request
#     payload = {
#         "model": model_name,
#         "messages": [
#             {
#                 "role": "user",
#                 "content": prompt
#             }
#         ],
#         "max_tokens": 300,
#         "temperature": 0.2
#     }
    
#     headers = {
#         'Authorization': f'Bearer {llm_key}',
#         'Content-Type': 'application/json'
#     }
    
#     # Convert payload to JSON string
#     json_payload = json.dumps(payload)
    
#     try:
#         # Send POST request to the API
#         response = requests.post(url, headers=headers, data=json_payload)
        
#         # Check if request was successful
#         if response.status_code == 200:            
#             response_data = response.json()                                 # Parse response JSON            
#             answer = response_data['choices'][0]['message']['content']      # Extract model's answer from response
#             return answer
#         else:
#             print(f"Request failed with status code: {response.status_code}")
#             return None
    
#     except requests.exceptions.RequestException as e:
#         print(f"Error occurred: {e}")
#         return None

# from huggingface_hub import InferenceClient

# def get_hugging_face_answer(model_name, context, question):
#     client = InferenceClient(model_name, token=os.getenv("HF_TOKEN"))
#     prompt = f"You are a Question Answering Model. Can you help me answer the question: {question} from the context: {context}? Just return the answer only. The document may contain some Arabic text; please translate that to English if needed."
#     output = client.text_generation(prompt , max_new_tokens = 200, stream=True, temperature=0.1)
#     return output

import os
from groq import Groq
from dotenv import load_dotenv
load_dotenv()

GROQ_API_KEY = os.getenv('GROQ_API')
def get_answer_from_context(model_name, context, question):
    client = Groq(api_key=GROQ_API_KEY)
    chat_completion = client.chat.completions.create(
        model=model_name,
        messages=[
            {
                "role": "system",
                "content": f"You are a Question Answering LLM that uses context provided: {context} to answer user's query. Just return the answer only. The document may contain some Arabic text; please translate that to English if needed."
            },
            {
                "role": "user",
                "content": question,
            }
        ],        
        temperature=0.2,
        max_tokens=200,
        top_p=1,
        stop=None,
        stream=False,
        # response_format={"type": "json_object"}
    )
    return chat_completion.choices[0].message.content