based on what? LLM benchmarks are all bullshit, so this is based on... your gut?

Gemini outputs what I want with a similar regularity as the other bots.

I'm so tired of the religious thinking around these models. show me a measurement.