e9d73ee30e
- Ollama 메인 → MLX fallback 순서로 변경 (기존 MLX 우선 제거) - OLLAMA_BASE_URL을 gpu.hyungi.net으로 변경 (Docker 네트워크 호환) - OLLAMA_TEXT_MODEL을 qwen3:8b → qwen3.5:9b-q8_0으로 업데이트 - health 엔드포인트: model 필드 직접 반환, 이중 중첩 해소 - health 체크 타임아웃 120초 → 5초로 단축 - Ollama API 호출에 think: false 추가 (thinking 토큰 방지) Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
31 lines
783 B
Python
31 lines
783 B
Python
from pydantic_settings import BaseSettings
|
|
|
|
|
|
class Settings(BaseSettings):
|
|
OLLAMA_BASE_URL: str = "https://gpu.hyungi.net"
|
|
OLLAMA_TEXT_MODEL: str = "qwen3.5:9b-q8_0"
|
|
OLLAMA_EMBED_MODEL: str = "bge-m3"
|
|
OLLAMA_TIMEOUT: int = 120
|
|
|
|
MLX_BASE_URL: str = "https://llm.hyungi.net"
|
|
MLX_TEXT_MODEL: str = "/Users/hyungi/mlx-models/Qwen3.5-27B-4bit"
|
|
|
|
DB_HOST: str = "mariadb"
|
|
DB_PORT: int = 3306
|
|
DB_USER: str = "hyungi_user"
|
|
DB_PASSWORD: str = ""
|
|
DB_NAME: str = "hyungi"
|
|
|
|
SECRET_KEY: str = ""
|
|
ALGORITHM: str = "HS256"
|
|
|
|
SYSTEM1_API_URL: str = "http://system1-api:3005"
|
|
CHROMA_PERSIST_DIR: str = "/app/data/chroma"
|
|
METADATA_DB_PATH: str = "/app/data/metadata.db"
|
|
|
|
class Config:
|
|
env_file = ".env"
|
|
|
|
|
|
settings = Settings()
|