All MicroEvals
Compare models
Create MicroEval

Compare models

Prompt

I want to compare models based on their cost per task and also their speed. I am looking at particularly latest opus 5.5, latest sonnet 5.5, latest sol-6, latest luna and gemini _8 flash. I have many agents one for explroting codebase, one for running tests (not luna), then my build and plan agents are both sol-6 medium and implement is gemini 3_8 flash (high) so build delegate things and lastly I have my reviewer which is opus 5.5 high. Question is theis the right models for the task based on compute per task anf type of task vs cost and also the correct effort levels

Drag to resize
Drag to resize
Drag to resize