Wyolet Relay – high throughput, open source LLM router
Self-hosted LLM gateway pooling keys across providers for failover and cost tracking.

One API key for Suno and Kling, but Replicate and Together AI already aggregate models.
Developers building applications that need multiple AI model integrations
Replicate · Together AI · Hugging Face Inference API
So I built RunAPI, which lets you call image, video, music/audio, and LLM models with one API key, and I also released SDKs, a CLI, skills, and an MCP server. This makes it easy for vibe coding to integrate these models into your app.
You can use it directly inside Claude Code, Codex, and even agents like OpenClaw and Hermes.
I’d love for you to try it and share feedback.
Self-hosted LLM gateway pooling keys across providers for failover and cost tracking.
Another LLM gateway when LiteLLM and Portkey already dominate the space.
OpenRouter clone with crypto payments in a well-served category.
Finally, binaural beats that don't clash musically with your actual playlist.
Terminal music player bypassing YouTube ads using yt-dlp and mpv.
Finally one CLI for Ollama, llama.cpp, and vLLM instead of three separate tools.