llm.py

import httpx
import asyncio
from config import CEREBRAS_API_KEY, GROQ_API_KEY, TIMEOUT_SECONDS, MAX_RETRIES, SYSTEM_PROMPT

async def fetch_completion(url: str, headers: dict, data: dict) -> str:
    async with httpx.AsyncClient(timeout=TIMEOUT_SECONDS) as client:
        response = await client.post(url, headers=headers, json=data)
        response.raise_for_status()
        return response.json()["choices"][0]["message"]["content"]

async def call_llm(task: str) -> str:
    messages = [
        {"role": "system", "content": SYSTEM_PROMPT},
        {"role": "user", "content": task}
    ]

    # Try Cerebras
    if CEREBRAS_API_KEY:
        try:
            return await fetch_completion(
                "https://api.cerebras.ai/v1/chat/completions",
                {"Authorization": f"Bearer {CEREBRAS_API_KEY}", "Content-Type": "application/json"},
                {"model": "llama-3.3-70b", "messages": messages}
            )
        except Exception as e:
            pass

    # Try Groq
    if GROQ_API_KEY:
        try:
            return await fetch_completion(
                "https://api.groq.com/openai/v1/chat/completions",
                {"Authorization": f"Bearer {GROQ_API_KEY}", "Content-Type": "application/json"},
                {"model": "llama-3.3-70b-versatile", "messages": messages}
            )
        except Exception as e:
            pass

    return "Error: All available models failed or no API keys provided."