File size: 2,887 Bytes
2b88732 718aa48 2b88732 fb07cb8 2b88732 fb07cb8 2b88732 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 |
import asyncio
import httpx
import os
OPENROUTER_BASE = "https://openrouter.ai/api/v1/chat/completions"
HEADERS = {
"Authorization": f"Bearer {os.getenv('OPENROUTER_API_KEY')}",
"Content-Type": "application/json",
}
ALLOWED_MODELS = [
"deepseek/deepseek-chat-v3-0324:free",
"google/gemini-2.0-flash-exp:free",
"meta-llama/llama-4-maverick:free",
"microsoft/mai-ds-r1:free",
"meta-llama/llama-4-scout:free",
"google/gemma-3-27b-it:free",
"qwen/qwq-32b:free",
"qwen/qwen2.5-vl-72b-instruct:free",
"qwen/qwen-2.5-72b-instruct:free",
"google/gemini-2.5-pro-exp-03-25:free",
"deepseek/deepseek-r1:free",
]
async def call_openrouter(model: str, prompt: str, system_prompt: str = None) -> str:
messages = []
if system_prompt:
messages.append({"role": "system", "content": system_prompt})
messages.append({"role": "user", "content": prompt})
body = {
"model": model,
"messages": messages,
"temperature": 0.7,
}
async with httpx.AsyncClient(timeout=30) as client:
response = await client.post(OPENROUTER_BASE, headers=HEADERS, json=body)
response.raise_for_status()
return response.json()["choices"][0]["message"]["content"]
async def query_llm_agent(name: str, prompt: str, settings: dict) -> str:
selected_model = settings.get("models", {}).get(name)
if not selected_model:
return f"[{name}] No model selected."
if selected_model not in ALLOWED_MODELS:
return f"[{name}] Model '{selected_model}' is not supported."
try:
response = await call_openrouter(selected_model, prompt)
return f"[{name}] {response}"
except Exception as e:
return f"[{name}] Error: {str(e)}"
async def query_all_llms(prompt: str, settings: dict) -> list:
agents = ["LLM-A", "LLM-B", "LLM-C"]
tasks = [query_llm_agent(agent, prompt, settings) for agent in agents]
results = await asyncio.gather(*tasks)
return results
def query_all_llms_sync(prompt: str, settings: dict) -> list:
return asyncio.run(query_all_llms(prompt, settings))
# Aggregator logic
async def query_aggregator(responses: list, settings: dict) -> str:
aggregator_model = settings.get("aggregator", "meta-llama/llama-4-scout:free")
if aggregator_model not in ALLOWED_MODELS:
return "[Aggregator] Model not allowed."
system_prompt = "You are an AI tasked with synthesizing the following responses from multiple language models into a single, coherent, high-quality answer."
joined_input = "\n\n".join(responses)
try:
return await call_openrouter(aggregator_model, joined_input, system_prompt)
except Exception as e:
return f"[Aggregator] Error: {str(e)}"
def aggregate_responses(responses: list, settings: dict) -> str:
return asyncio.run(query_aggregator(responses, settings))
|