Collected sources and patterns will appear here. Add from search, explore, or the patterns library.
Prompt -> Map<ModelName, Response>
Dispatch a single prompt to multiple downstream LLM APIs concurrently to aggregate and compare their outputs.
Problem it solves
Comparing model outputs requires manual copy-pasting across multiple web interfaces.
Consumes
Emits
The real projects this mechanism was found in. Attribution is the point — this is how the best teams actually do it.