29 lines
1003 B
Python
29 lines
1003 B
Python
|
import os
|
||
|
|
from pathlib import Path
|
||
|
|
|
||
|
|
from dotenv import load_dotenv
|
||
|
|
|
||
|
|
load_dotenv(Path(__file__).resolve().parent.parent / ".env", override=True)
|
||
|
|
|
||
|
|
GEMINI_API_KEY = os.environ.get("GEMINI_API_KEY", "")
|
||
|
|
GEMINI_MODEL = os.environ.get("GEMINI_MODEL", "gemini-2.5-pro")
|
||
|
|
GEMINI_URL = (
|
||
|
|
f"https://generativelanguage.googleapis.com/v1beta/models/{GEMINI_MODEL}:generateContent"
|
||
|
|
)
|
||
|
|
|
||
|
|
# Ollama (local VLM — default)
|
||
|
|
OLLAMA_BASE_URL = os.environ.get("OLLAMA_BASE_URL", "http://localhost:11434")
|
||
|
|
OLLAMA_MODEL = os.environ.get("OLLAMA_MODEL", "qwen3.5:9b")
|
||
|
|
|
||
|
|
# VLM backend: "ollama" (default) or "gemini"
|
||
|
|
VLM_BACKEND = os.environ.get("VLM_BACKEND", "ollama")
|
||
|
|
|
||
|
|
BACKEND_BASE_URL = os.environ.get("BACKEND_BASE_URL", "https://wahwa.com/api/v1")
|
||
|
|
BACKEND_JWT = os.environ.get("BACKEND_JWT", "")
|
||
|
|
|
||
|
|
CAPTURE_INTERVAL_S = 2.5 # screenshot every 2.5s
|
||
|
|
BUFFER_MAX_LEN = 4 # 4 frames accumulated
|
||
|
|
VLM_CALL_EVERY_N = 4 # call VLM every 4 captures (= every 10s)
|
||
|
|
SCREENSHOT_JPEG_QUALITY = 50
|
||
|
|
SCREENSHOT_MAX_WIDTH = 1280
|