Configuration¶

All configuration options for Savanty.

Environment Variables¶

Required¶

Variable	Description
`OPENAI_API_KEY`	Your OpenAI API key for LLM access

LLM Configuration¶

Variable	Default	Description
`SAVANTY_LLM_MODEL`	`openai/gpt-4o`	LLM model to use

Supported models:

openai/gpt-4o (recommended)
openai/gpt-4-turbo
openai/gpt-3.5-turbo (faster, less accurate)

Server Configuration¶

Variable	Default	Description
`SAVANTY_PORT`	`8000`	Web server port
`SAVANTY_ENV`	`development`	Environment mode
`SAVANTY_CORS_ORIGINS`	localhost	Allowed CORS origins

CORS origins format:

Bash

# Single origin
SAVANTY_CORS_ORIGINS=https://myapp.com

# Multiple origins (comma-separated)
SAVANTY_CORS_ORIGINS=https://myapp.com,https://api.myapp.com

Request Limits¶

Variable	Default	Description
`SAVANTY_MAX_PROBLEM_LENGTH`	`10000`	Max chars in problem description
`SAVANTY_MAX_ADDITIONAL_INFO_LENGTH`	`5000`	Max chars in additional info
`SAVANTY_SOLVE_TIMEOUT`	`120`	Solve timeout in seconds

Logging¶

Variable	Default	Description
`SAVANTY_LOG_LEVEL`	`INFO`	Log level (DEBUG, INFO, WARNING, ERROR)
`SAVANTY_LOG_FORMAT`	See below	Log message format

Default log format:

Text Only

%(asctime)s - %(name)s - %(levelname)s - %(message)s

Configuration File¶

Create a .env file in your project root:

Bash

# .env
OPENAI_API_KEY=sk-...

# Optional
SAVANTY_LLM_MODEL=openai/gpt-4o
SAVANTY_PORT=8000
SAVANTY_LOG_LEVEL=INFO

Load with python-dotenv:

Python

from dotenv import load_dotenv
load_dotenv()

from savanty import solve_optimization_problem

Example Configurations¶

Development¶

Bash

export OPENAI_API_KEY=sk-...
export SAVANTY_ENV=development
export SAVANTY_LOG_LEVEL=DEBUG

Production¶

Bash

export OPENAI_API_KEY=sk-...
export SAVANTY_ENV=production
export SAVANTY_CORS_ORIGINS=https://yourdomain.com
export SAVANTY_LOG_LEVEL=WARNING
export SAVANTY_SOLVE_TIMEOUT=60

High-Volume¶

Bash

export OPENAI_API_KEY=sk-...
export SAVANTY_LLM_MODEL=openai/gpt-3.5-turbo  # Faster
export SAVANTY_SOLVE_TIMEOUT=30
export SAVANTY_MAX_PROBLEM_LENGTH=5000

Programmatic Configuration¶

Set environment variables before importing:

Python

import os
os.environ["OPENAI_API_KEY"] = "sk-..."
os.environ["SAVANTY_LLM_MODEL"] = "openai/gpt-4o"

from savanty import solve_optimization_problem

Configuration Precedence¶

Environment variables (highest priority)
.env file (if using python-dotenv)
Default values (lowest priority)