Check quality and compare options
Measure whether a prompt, model or workflow is good enough before and after you change it.
2 entries
-
FM-04
Build a small evaluation set before you change a prompt
You end with A versioned file of test cases, a baseline score for the current prompt, and a per-case comparison for each change.
Evaluate · Engineering
- Level
- Practitioner
- Effort
- 1 h 45 min
- Works in
- Model API and scripts, full support
- Coding agent in a repository, full support
- Spreadsheet, partial support
- Chat assistant, partial support
- Record
- Reviewed · v1.1
-
FM-05
Compare two models or prompts blind on your own task
You end with A tally of wins, losses and ties per criterion, and a decision with the tie rule written down in advance.
Evaluate · Research
- Level
- Practitioner
- Effort
- 1 h 10 min
- Works in
- Model API and scripts, full support
- Spreadsheet, full support
- Chat assistant, partial support
- Record
- Reviewed · v1.0