Configuration MoneyPrinter reads configuration from a .env file in the project root. Use .env.example as your starting template.
Environment File Create your configuration file:
Edit .env with your preferred text editor:
# Required API Keys TIKTOK_SESSION_ID = "" PEXELS_API_KEY = "" # Optional Configuration IMAGEMAGICK_BINARY = "" OLLAMA_BASE_URL = "http://localhost:11434" OLLAMA_MODEL = "llama3.1:8b" OLLAMA_TIMEOUT = "180" ASSEMBLY_AI_API_KEY = "" # Database (Docker) POSTGRES_DB = "moneyprinter" POSTGRES_USER = "moneyprinter" POSTGRES_PASSWORD = "moneyprinter" DATABASE_URL = ""
Required Variables TIKTOK_SESSION_ID Description : TikTok session cookie used for text-to-speech API calls.How to obtain :
Open Developer Tools
Press F12 or right-click and select “Inspect Element”.
Navigate to Cookies
Go to Application → Cookies → https://www.tiktok.com.
Copy sessionid
Find the sessionid cookie and copy its value.
TIKTOK_SESSION_ID = "abc123def456..."
The session ID may expire periodically. If TTS fails with authentication errors, refresh this value.
PEXELS_API_KEY Description : API key for fetching stock video clips from Pexels.How to obtain :
Generate API key
Navigate to your API dashboard and generate a new key.
PEXELS_API_KEY = "your_pexels_api_key_here"
Pexels offers a generous free tier (200 requests/hour). No credit card required.
Optional Variables IMAGEMAGICK_BINARY Description : Absolute path to the ImageMagick executable.Default : Auto-detect from PATHThe backend auto-detects ImageMagick on startup:
def resolve_imagemagick_binary () -> Optional[ str ]: configured_binary = os.getenv( "IMAGEMAGICK_BINARY" , "" ).strip().strip( '"' ) if configured_binary: expanded = Path(configured_binary).expanduser() if expanded.exists(): return str (expanded.resolve()) candidate_names = [ "magick" , "magick.exe" , "convert" , "convert.exe" , ] for candidate in candidate_names: found = shutil.which(candidate) if found: return found return None
If auto-detection fails, set the path explicitly:
IMAGEMAGICK_BINARY = "/usr/local/bin/magick"
Windows paths must use double backslashes (\\) or forward slashes (/).
OLLAMA_BASE_URL Description : Base URL of the Ollama server for model listing and chat generation.Default : http://localhost:11434Used by the Ollama client:
OLLAMA_BASE_URL = os.getenv( "OLLAMA_BASE_URL" , "http://localhost:11434" ).rstrip( "/" ) def _ollama_client () -> Client: return Client( host = OLLAMA_BASE_URL , timeout = OLLAMA_TIMEOUT )
Examples :# Local Ollama OLLAMA_BASE_URL = "http://localhost:11434" # Docker backend connecting to host Ollama OLLAMA_BASE_URL = "http://host.docker.internal:11434" # Remote Ollama server OLLAMA_BASE_URL = "http://192.168.1.100:11434"
OLLAMA_MODEL Description : Fallback model name if the frontend doesn’t specify one.Default : llama3.1:8bOLLAMA_MODEL = os.getenv( "OLLAMA_MODEL" , "llama3.1:8b" ) def generate_response ( prompt : str , ai_model : str ) -> str : model_name = (ai_model or "" ).strip() or OLLAMA_MODEL # ...
Pull the model before use:
Recommended models:
Model Size Speed Quality Use Case llama3.1:8b4.7GB Fast Good General purpose mistral:7b4.1GB Very Fast Good Quick iterations llama3.1:70b40GB Slow Excellent Best quality (GPU required) qwen2.5:7b4.7GB Fast Good Multilingual
OLLAMA_TIMEOUT Description : Timeout in seconds for Ollama API requests.Default : 180 (3 minutes)OLLAMA_TIMEOUT = float (os.getenv( "OLLAMA_TIMEOUT" , "180" )) def _ollama_client () -> Client: return Client( host = OLLAMA_BASE_URL , timeout = OLLAMA_TIMEOUT )
Increase this value if you’re using large models that take longer to respond:
# For slower models or resource-constrained systems OLLAMA_TIMEOUT = 300
Large models (70B+) may require longer timeouts, especially on CPU-only systems.
ASSEMBLY_AI_API_KEY Description : API key for AssemblyAI subtitle generation. If empty, local subtitle generation is used.Default : Empty (use local processing)ASSEMBLY_AI_API_KEY = "your_assemblyai_key_here"
Local vs AssemblyAI subtitles:
Feature Local AssemblyAI Cost Free Paid ($0.00025/sec) Quality Basic High accuracy Speed Fast Network latency Privacy Fully local Cloud upload
For most use cases, local subtitle generation is sufficient. AssemblyAI provides better accuracy for complex audio.
Database Configuration DATABASE_URL Description : SQLAlchemy connection string for the job queue database.Default : sqlite:///moneyprinter.db (local file)The database is initialized on startup:
import os from sqlalchemy import create_engine from sqlalchemy.orm import declarative_base, sessionmaker DATABASE_URL = os.getenv( "DATABASE_URL" , "sqlite:///moneyprinter.db" ) engine = create_engine( DATABASE_URL ) SessionLocal = sessionmaker( bind = engine) Base = declarative_base()
Examples :# Local SQLite (default) DATABASE_URL = "sqlite:///moneyprinter.db" # Docker Postgres DATABASE_URL = "postgresql+psycopg://moneyprinter:moneyprinter@postgres:5432/moneyprinter" # External Postgres DATABASE_URL = "postgresql+psycopg://user:pass@host:5432/dbname"
POSTGRES_DB, POSTGRES_USER, POSTGRES_PASSWORD Description : Docker Compose Postgres service configuration.Defaults :
POSTGRES_DB=moneyprinterPOSTGRES_USER=moneyprinterPOSTGRES_PASSWORD=moneyprinter
Used by docker-compose.yml:
services : postgres : image : postgres:16-alpine environment : - POSTGRES_DB=${POSTGRES_DB:-moneyprinter} - POSTGRES_USER=${POSTGRES_USER:-moneyprinter} - POSTGRES_PASSWORD=${POSTGRES_PASSWORD:-moneyprinter}
Production : Change POSTGRES_PASSWORD to a strong, unique password.
Validation The backend validates required variables on startup:
def check_env_vars () -> None : required_vars = [ "PEXELS_API_KEY" , "TIKTOK_SESSION_ID" ] missing_vars = [] for var in required_vars: value = os.getenv(var) if value is None or len (value) == 0 : missing_vars.append(var) if missing_vars: logger.error( f "Missing environment variables: { ', ' .join(missing_vars) } " ) sys.exit( 1 ) # Verify ImageMagick imagemagick_binary = resolve_imagemagick_binary() if not imagemagick_binary: logger.error( "ImageMagick not found" ) sys.exit( 1 )
If variables are missing, the backend exits with an error:
ERROR: The following environment variables are missing: PEXELS_API_KEY, TIKTOK_SESSION_ID Please consult 'docs/configuration.md' for setup instructions.
Next Steps
Quickstart Generate your first video
Docker Setup Deploy with Docker Compose
Ollama Models Install and select Ollama models
Troubleshooting Common configuration issues