Low-Latency Model Router: Automatic LLM Selection Across OpenRouter

When calling an LLM API directly, the model selection is typically fixed ahead of time. In practice,...