Show HN: Swobu – local TUI switchboard for LLM you can share over E2E HTTPS (github.com)

Swobu has emerged as a revolutionary tool for managing local LLM (Large Language Model) resources, allowing users to create a centralized switchboard that can route requests across various AI model providers like Claude Code and Codex. This innovative approach enables the pooling of existing model capacities from various sources, such as AWS or Azure credits and free APIs, behind stable HTTPS routes. With Swobu, users can seamlessly change model configurations and allocate resources without needing to reconfigure each client individually, streamlining the interaction with LLMs. This development is significant for the AI/ML community as it simplifies resource management and enhances the utilization of heterogeneous AI resources. Swobu supports 40 different provider integrations, allowing for consistent performance across varied environments. Its design emphasizes privacy and security by maintaining routing details and credentials server-side, thus preventing exposure to client applications. Swobu’s flexibility includes automatic fallback to alternative models when primary options fail, ensuring uninterrupted service and efficient capacity management. Overall, Swobu stands out as a practical solution for developers seeking to optimize their AI operations while minimizing complications from managing multiple models and accounts.