When you call an LLM API, you pick a model and hope it's the right one. Claude for reasoning, GPT for code, Gemini for long context... you're doing the routing manually.

Source: [Dev.to](https://dev.to/moozechen/i-built-an-ai-router-that-decomposes-your-prompt-and-dispatches-to-the-best-model-agc)

Sponsored