Hapax Documentation
The reliability layer between your code and LLM providers
A lightweight, production-ready infrastructure layer that ensures continuous operation through intelligent provider management and automatic failover.
Why Hapax?
Hapax addresses the fundamental challenges of working with LLM providers: service reliability, provider management, and operational visibility.
Key Benefits
Continuous Operation
Automatic failover between providers maintains service availability during outages or degraded performance.
Minimal Configuration
Single configuration file handles all provider settings, health checks, and failover logic.
Operational Insight
Built-in metrics expose detailed provider performance, request patterns, and system health.
Core Features
Intelligent Provider Management
- Health monitoring with configurable thresholds
- Automatic provider failover
- Circuit breaker implementation
- Request deduplication
System Architecture
- Request routing and load distribution
- Comprehensive error handling
- Structured logging with request tracing
- HTTP/3 support
Security and Monitoring
- API key-based authentication
- Request validation
- Usage metrics per endpoint
- Prometheus integration
Documentation
Development
Find issues or want to contribute?