mirror of
https://github.com/x1xhlol/system-prompts-and-models-of-ai-tools.git
synced 2026-06-18 07:19:35 +00:00
220 lines
8.8 KiB
Python
220 lines
8.8 KiB
Python
"""
|
|
Multi-Model AI Router — Saudi AI Company Brain
|
|
Routes requests to the best AI model based on task type.
|
|
GLM-5 = Sales Brain | Groq = Fast Classification | Claude = Copy | Gemini = Research
|
|
"""
|
|
import httpx
|
|
import json
|
|
import os
|
|
import logging
|
|
from typing import Optional, Dict, Any
|
|
|
|
logger = logging.getLogger(__name__)
|
|
|
|
|
|
class ModelRouter:
|
|
"""Routes AI requests to the optimal model based on task type."""
|
|
|
|
ROUTING_TABLE = {
|
|
# GLM-5 (Z.AI) — Sales decisions, closing, strategy
|
|
"sales_decision": "glm5",
|
|
"close": "glm5",
|
|
"strategy": "glm5",
|
|
"followup_plan": "glm5",
|
|
"lead_qualify": "glm5",
|
|
"objection_handle": "glm5",
|
|
|
|
# Groq — Fast classification, tagging
|
|
"fast_classify": "groq",
|
|
"lead_score": "groq",
|
|
"intent_detect": "groq",
|
|
"sentiment": "groq",
|
|
"tag": "groq",
|
|
|
|
# Claude — Copy, proposals, landing pages
|
|
"proposal_copy": "claude",
|
|
"landing_copy": "claude",
|
|
"email_draft": "claude",
|
|
"whatsapp_template": "claude",
|
|
"report": "claude",
|
|
|
|
# Gemini — Research, document analysis
|
|
"research": "gemini",
|
|
"document_analysis": "gemini",
|
|
"market_intel": "gemini",
|
|
"competitor_analysis": "gemini",
|
|
|
|
# DeepSeek — Code, integrations
|
|
"coding": "deepseek",
|
|
"integration": "deepseek",
|
|
"debug": "deepseek",
|
|
}
|
|
|
|
def __init__(self):
|
|
self.groq_key = os.getenv("GROQ_API_KEY", "")
|
|
self.anthropic_key = os.getenv("ANTHROPIC_API_KEY", "")
|
|
self.deepseek_key = os.getenv("DEEPSEEK_API_KEY", "")
|
|
self.zai_key = os.getenv("ZAI_API_KEY", "")
|
|
self.gemini_key = os.getenv("GOOGLE_API_KEY", "")
|
|
self.zai_base = os.getenv("ZAI_BASE_URL", "https://api.z.ai/api/paas/v4/")
|
|
|
|
def get_model_for_task(self, task_type: str) -> str:
|
|
return self.ROUTING_TABLE.get(task_type, "groq")
|
|
|
|
async def route(self, task_type: str, prompt: str,
|
|
system_prompt: str = "", temperature: float = 0.3,
|
|
max_tokens: int = 2048) -> Dict[str, Any]:
|
|
"""Route request to the best model."""
|
|
model_id = self.get_model_for_task(task_type)
|
|
|
|
try:
|
|
if model_id == "groq":
|
|
return await self._call_groq(prompt, system_prompt, temperature, max_tokens)
|
|
elif model_id == "glm5":
|
|
return await self._call_glm5(prompt, system_prompt, temperature, max_tokens)
|
|
elif model_id == "claude":
|
|
return await self._call_claude(prompt, system_prompt, temperature, max_tokens)
|
|
elif model_id == "gemini":
|
|
return await self._call_gemini(prompt, system_prompt, temperature, max_tokens)
|
|
elif model_id == "deepseek":
|
|
return await self._call_deepseek(prompt, system_prompt, temperature, max_tokens)
|
|
else:
|
|
return await self._call_groq(prompt, system_prompt, temperature, max_tokens)
|
|
except Exception as e:
|
|
logger.warning(f"Model {model_id} failed: {e}, falling back to Groq")
|
|
try:
|
|
return await self._call_groq(prompt, system_prompt, temperature, max_tokens)
|
|
except Exception as e2:
|
|
return {"text": f"All models failed: {e2}", "model": "error", "error": True}
|
|
|
|
async def _call_groq(self, prompt: str, system: str = "",
|
|
temp: float = 0.3, max_tokens: int = 2048) -> Dict:
|
|
if not self.groq_key:
|
|
return {"text": "GROQ_API_KEY not set", "model": "groq", "error": True}
|
|
|
|
messages = []
|
|
if system:
|
|
messages.append({"role": "system", "content": system})
|
|
messages.append({"role": "user", "content": prompt})
|
|
|
|
async with httpx.AsyncClient(timeout=30) as client:
|
|
resp = await client.post(
|
|
"https://api.groq.com/openai/v1/chat/completions",
|
|
headers={"Authorization": f"Bearer {self.groq_key}"},
|
|
json={
|
|
"model": os.getenv("GROQ_MODEL", "llama-3.3-70b-versatile"),
|
|
"messages": messages,
|
|
"temperature": temp,
|
|
"max_tokens": max_tokens,
|
|
}
|
|
)
|
|
data = resp.json()
|
|
return {
|
|
"text": data["choices"][0]["message"]["content"],
|
|
"model": "groq",
|
|
"usage": data.get("usage", {}),
|
|
}
|
|
|
|
async def _call_glm5(self, prompt: str, system: str = "",
|
|
temp: float = 0.3, max_tokens: int = 2048) -> Dict:
|
|
if not self.zai_key:
|
|
logger.warning("ZAI_API_KEY not set, falling back to Groq")
|
|
return await self._call_groq(prompt, system, temp, max_tokens)
|
|
|
|
messages = []
|
|
if system:
|
|
messages.append({"role": "system", "content": system})
|
|
messages.append({"role": "user", "content": prompt})
|
|
|
|
async with httpx.AsyncClient(timeout=30) as client:
|
|
resp = await client.post(
|
|
f"{self.zai_base}chat/completions",
|
|
headers={"Authorization": f"Bearer {self.zai_key}"},
|
|
json={
|
|
"model": "glm-4-plus",
|
|
"messages": messages,
|
|
"temperature": temp,
|
|
"max_tokens": max_tokens,
|
|
}
|
|
)
|
|
data = resp.json()
|
|
text = data.get("choices", [{}])[0].get("message", {}).get("content", "")
|
|
return {"text": text, "model": "glm5", "usage": data.get("usage", {})}
|
|
|
|
async def _call_claude(self, prompt: str, system: str = "",
|
|
temp: float = 0.3, max_tokens: int = 2048) -> Dict:
|
|
if not self.anthropic_key:
|
|
return await self._call_groq(prompt, system, temp, max_tokens)
|
|
|
|
async with httpx.AsyncClient(timeout=60) as client:
|
|
resp = await client.post(
|
|
"https://api.anthropic.com/v1/messages",
|
|
headers={
|
|
"x-api-key": self.anthropic_key,
|
|
"anthropic-version": "2023-06-01",
|
|
"content-type": "application/json",
|
|
},
|
|
json={
|
|
"model": "claude-sonnet-4-20250514",
|
|
"max_tokens": max_tokens,
|
|
"system": system or "You are a Saudi AI sales expert.",
|
|
"messages": [{"role": "user", "content": prompt}],
|
|
}
|
|
)
|
|
data = resp.json()
|
|
text = data.get("content", [{}])[0].get("text", "")
|
|
return {"text": text, "model": "claude", "usage": data.get("usage", {})}
|
|
|
|
async def _call_gemini(self, prompt: str, system: str = "",
|
|
temp: float = 0.3, max_tokens: int = 2048) -> Dict:
|
|
if not self.gemini_key:
|
|
return await self._call_groq(prompt, system, temp, max_tokens)
|
|
|
|
full_prompt = f"{system}\n\n{prompt}" if system else prompt
|
|
async with httpx.AsyncClient(timeout=30) as client:
|
|
resp = await client.post(
|
|
f"https://generativelanguage.googleapis.com/v1beta/models/gemini-2.0-flash:generateContent?key={self.gemini_key}",
|
|
json={
|
|
"contents": [{"parts": [{"text": full_prompt}]}],
|
|
"generationConfig": {"temperature": temp, "maxOutputTokens": max_tokens}
|
|
}
|
|
)
|
|
data = resp.json()
|
|
text = data.get("candidates", [{}])[0].get("content", {}).get("parts", [{}])[0].get("text", "")
|
|
return {"text": text, "model": "gemini", "usage": {}}
|
|
|
|
async def _call_deepseek(self, prompt: str, system: str = "",
|
|
temp: float = 0.3, max_tokens: int = 2048) -> Dict:
|
|
if not self.deepseek_key:
|
|
return await self._call_groq(prompt, system, temp, max_tokens)
|
|
|
|
messages = []
|
|
if system:
|
|
messages.append({"role": "system", "content": system})
|
|
messages.append({"role": "user", "content": prompt})
|
|
|
|
async with httpx.AsyncClient(timeout=30) as client:
|
|
resp = await client.post(
|
|
"https://api.deepseek.com/chat/completions",
|
|
headers={"Authorization": f"Bearer {self.deepseek_key}"},
|
|
json={
|
|
"model": "deepseek-chat",
|
|
"messages": messages,
|
|
"temperature": temp,
|
|
"max_tokens": max_tokens,
|
|
}
|
|
)
|
|
data = resp.json()
|
|
text = data["choices"][0]["message"]["content"]
|
|
return {"text": text, "model": "deepseek", "usage": data.get("usage", {})}
|
|
|
|
|
|
# Singleton
|
|
_router: Optional[ModelRouter] = None
|
|
|
|
def get_router() -> ModelRouter:
|
|
global _router
|
|
if _router is None:
|
|
_router = ModelRouter()
|
|
return _router
|