.env.example

HF_EMBEDDING_MODEL=BAAI/bge-small-en-v1.5
CHUNK_SIZE=512
CHUNK_OVERLAP=64
TOP_K=4
FAISS_INDEX_PATH=rag/faiss_index
MAX_HISTORY=12
AUTO_BUILD_INDEX=false

# Local Hugging Face LLM (no OpenRouter)
# 4GB VRAM profile:
# HF_LLM_MODEL=Qwen/Qwen2.5-1.5B-Instruct
# HF_LLM_LOAD_IN_4BIT=false
#
# 16GB VRAM profile (recommended for better quality):
# HF_LLM_MODEL=Raj-Maharajwala/Open-Insurance-LLM-Llama3-8B
# HF_LLM_LOAD_IN_4BIT=true
HF_LLM_MODEL=Qwen/Qwen2.5-1.5B-Instruct
HF_LLM_DEVICE=auto
HF_LLM_DTYPE=auto
HF_LLM_LOAD_IN_4BIT=false
HF_LLM_MAX_NEW_TOKENS=512
HF_LLM_TEMPERATURE=0.1
HF_LLM_TOP_P=0.9
HF_LLM_REPETITION_PENALTY=1.05