MMLU-Pro is named by AI engines in 3 questions across 1 segment.
5 observations3 of 5 engines name itLast seen Sep 9, 2026
Where it is named
| Question | Avg position | Named in | Engines |
|---|---|---|---|
| What are the top AI evaluation & benchmarks tools in 2026? AI Evaluation | #1.0 | 1 of 12 answers | Gemini |
| What is the best AI benchmark test? AI Evaluation | #3.5 ±0.7 | 2 of 5 answers | ChatGPT, Copilot |
| What is the best AI performance benchmark? AI Evaluation | #6.0 ±2.8 | 2 of 5 answers | ChatGPT, Copilot |
Which engines name it — and which do not
- ChatGPT2 questions
- Copilot2 questions
- Gemini1 question
- Google AI Modenever named
- Perplexitynever named
ChatGPT names MMLU-Pro in 2 questions, Copilot names MMLU-Pro in 2 questions and Gemini names MMLU-Pro in 1 question. Google AI Mode, Perplexity have not named MMLU-Pro in any of these questions — that is where the ground is open.
Counted from the questions above. A question may not have run on every engine.
Where it is strongest and weakest
Strongest
Fewer than four ranked questions — no separate weakest set.
± is the spread of positions across answers. A wide spread means the ranking is unstable — which means it is winnable.
Source: ranking answers from 5 AI engines, first seen Sep 7, 2026, last seen Sep 9, 2026. This page is generated from observations only — it is not claimed or edited by MMLU-Pro.