Model Testing - Mode for Prompt Evaluation Across Models

I’d love a “Model Testing” feature in AI Flow Chat that enables me to run parallel model comparisons directly within an automation node. The idea is to split a single prompt across 2–5 selected models and see the output side by side.

This would be particularly useful when fine-tuning automations, as it would allow me to:

• Compare model outputs instantly

• Spot quality differences and consistency

• Identify the best-fitting model for a specific task

Key UX request:
This should ideally be surfaced as a dedicated modal or popup, where I can:

• Select which models to test

• Run the test multiple times (e.g., 3–5) for consistency checks

• Review and compare outputs in a clear, side-by-side view

• Finalize my model choice with confidence

This would turn model selection from guesswork into an empirical, workflow-integrated process.