Compare Two Models on My Own Tasks
Run a fair side-by-side test using your real work, not benchmarks.
Added October 15, 2026. Written with AI assistance and checked for clarity and placeholders; not reviewed by a lawyer, accountant, clinician or other professional. How these prompts are made
The Open buttons send the prompt text as currently shown below to that site only when you click. Fill in the placeholders first.
Design a fair comparison of two AI models on my work. Tasks: [TASK_LIST]. Models: [MODEL_A] and [MODEL_B]. Quality criteria: [CRITERIA]. Time I can spend: [HOURS]. Pick eight representative tasks from my list, including two hard ones. Use identical prompts and settings for both. Hide which model wrote each answer when scoring. Score each answer 1 to 5 per criterion and total them, showing the arithmetic. Record cost and time per answer. Return the test plan, a scoring sheet layout, rules to avoid bias such as testing in random order, and a short decision note format: winner per task type, overall winner, and what would change my mind. Note that results are specific to these tasks. Also record any task where both models failed, since that points to a prompt or data problem rather than a model difference, and keep the full set of prompts and answers so the test can be repeated when new versions appear.
Fill in before use
Optional: type your details and the prompt above updates. Nothing is saved or sent.
Replace each bracketed placeholder with your own details, then paste into Any AI or another assistant.