9router just got auto vision + audio model routing.
Configure it once. Any non-multimodal model called through 9router now detects vision/audio tasks and silently swaps to a combo model behind the scenes.
Reminds me of the vision subagent approach I tried before but this solves it at the API layer. No glue code, no orchestration headaches.
My take: models will keep getting narrower. The winning product is multiple specialists stitched together, not one jack-of-all-trades model.
