app.py

from fastapi import FastAPI, HTTPException
from pydantic import BaseModel
import os
import logging
import openai

# Read the NVIDIA API key from environment variables
api_key = os.getenv("NVIDIA_API_KEY")
if api_key is None:
    raise ValueError("NVIDIA API key not found in environment variables. Please set the NVIDIA_API_KEY.")

# Initialize FastAPI app
app = FastAPI()

# Set up logging
logging.basicConfig(level=logging.INFO)
logger = logging.getLogger(__name__)

# NVIDIA API configuration
openai.api_key = api_key
openai.base_url = "https://integrate.api.nvidia.com/v1"

# Define request body schema
class TextGenerationRequest(BaseModel):
    prompt: str
    max_new_tokens: int = 1024
    temperature: float = 0.4
    top_p: float = 0.7
    stream: bool = True

# Define API endpoint
@app.post("/generate-text")
async def generate_text(request: TextGenerationRequest):
    try:
        logger.info("Generating text...")

        # Generate response from NVIDIA API
        response = openai.ChatCompletion.create(
            model="meta/llama-3.1-405b-instruct",
            messages=[{"role": "user", "content": request.prompt}],
            temperature=request.temperature,
            top_p=request.top_p,
            max_tokens=request.max_new_tokens,
            stream=request.stream,
        )

        response_text = ""
        if request.stream:
            for chunk in response:
                if chunk.choices[0].delta.get("content"):
                    response_text += chunk.choices[0].delta.content
        else:
            response_text = response["choices"][0]["message"]["content"]

        return {"generated_text": response_text}
    except Exception as e:
        logger.error(f"Error generating text: {e}")
        raise HTTPException(status_code=500, detail=str(e))

# Add a root endpoint for health checks
@app.get("/")
async def root():
    return {"message": "Welcome Fred808 GPT"}

# Add a test endpoint
@app.get("/test")
async def test():
    return {"message": "API is running!"}