Scores each criterion from its own evidence in the answer
A raw score per criterion, on its scale, from the evidence for that criterion only
Can AI do this?
AI does this well Most of the tested AI models master this. 3 of 4 models in the current benchmark master it.
| Model | Tier | Probability of mastery | Status |
|---|---|---|---|
| gemini-4-argon | Flagship | 95% | Mastered |
| gemini-3.5-flash-lite | Low cost | 94% | Mastered |
| gpt-6-luna | Low cost | 18% | Not shown |
| claude-haiku-5-5 | Low cost | 95% | Mastered |
Mastered means a probability of at least 90% that the model has this knobit, estimated from its graded answers. “Not shown” means the answers do not settle it either way. How the models are tested.
What it looks like in practice
Expected result: A raw score per criterion, on its scale, from the evidence for that criterion only
How it is assessed: The scores agree with grades known to be correct
Related knobits
Builds on
All 16 knobits of Grade answers against a rubric.
Source: Graded on its own by sa-answer-grading