Configuration Reference

CLI Options

fallbackrabbit serve

Flag	Default	Description
--host	127.0.0.1	Bind address
--port	8000	Port number
--storage	memory://	Storage URL (memory:// or sqlite:///path.db)
--reload	false	Enable auto-reload for development
--api-keys	—	API keys (comma-separated or label:key pairs)
--rate-limit	60	Per-IP rate limit per minute
--rate-limit-global	600	Global rate limit per minute
--rate-limit-burst	10	Burst allowance

fallbackrabbit test

Flag	Default	Description
--prompts	5	Number of test prompts
--outages	—	Outage scenario file
--verbose	false	Detailed per-prompt results
--real	false	Make real provider calls

fallbackrabbit export

Flag	Default	Description
--format	litellm	Export format
--output	stdout	Output file path
--template	—	Jinja2 template file
--extra-vars	—	Template variables (key=value, repeatable)

Environment Variables

Variable	Description
OPENAI_API_KEY	OpenAI API key
ANTHROPIC_API_KEY	Anthropic API key
AZURE_OPENAI_API_KEY	Azure OpenAI API key
AZURE_OPENAI_ENDPOINT	Azure OpenAI endpoint URL
OLLAMA_BASE_URL	Ollama base URL (default: http://localhost:11434)
FALLBACKRABBIT_API_KEYS	API keys for authentication
FALLBACKRABBIT_STORAGE_URL	Default storage URL

Export Formats

Format	Description
litellm	LiteLLM model config
openrouter	OpenRouter routing config
langchain	LangChain router config
haystack	Haystack pipeline config
custom	Raw JSON dump
template	Jinja2 template-based

Error Types

Type	HTTP Codes	Default Action
rate_limit	429	Retry
timeout	408, 504	Failover
server_error	500, 502, 503	Failover
auth_error	401, 403	Fail
connection_error	—	Failover
validation_error	400	Fail

Fallback Actions

Action	Behavior
retry	Retry same provider (up to max_retries)
failover	Try next provider
wait	Wait wait_ms then retry same provider
fail	Stop immediately