config.py

# config.py
import os
from dotenv import load_dotenv

load_dotenv()

# models and paths
LLM_HF_MODEL = os.getenv("LLM_HF_MODEL", "./tokenizer/Qwen2.5-1.5B-Instruct") # 只需要根据实际路径修改这里的模型路径即可
LLM_AX_MODEL = os.getenv("LLM_AX_MODEL", "./models/Qwen2.5-1.5B-Instruct_axmodel")
EMBED_HF_MODEL = os.getenv("EMBED_HF_MODEL", "./tokenizer/Qwen3-Embedding-0.6B")
EMBED_AX_MODEL = os.getenv("EMBED_AX_MODEL", "./models/Qwen3-Embedding-0.6B_axmodel")

# API URL
LLM_API_PORT = int(os.getenv("LLM_API_PORT", "8000"))
LLM_API_URL = os.getenv("LLM_API_URL", f"http://127.0.0.1:{LLM_API_PORT}/generate") # 具体函数名字需要对应修改
PORT = int(os.getenv("PORT", "7860"))

# Index paths
INDEX_DIR = os.getenv("INDEX_DIR", "index")
INDEX_FILE = os.path.join(INDEX_DIR, "docs.index")
EMBEDDINGS_FILE = os.path.join(INDEX_DIR, "docs.pkl")

os.makedirs(INDEX_DIR, exist_ok=True)