benchmarks.wiki / Public workspace

MMLU-Pro

Reasoning

MMLU-Pro rebuilds MMLU with ten answer options instead of four and removes items that no longer separate models. Questions span fourteen academic and professional subjects.

mmlu-pro

Full text12,032 tasks

Official source

Full text: eligible public problems include text, images and discussion.

History

TIGER-Lab introduced MMLU-Pro in 2024 to make broad academic question answering more demanding as the original MMLU became less discriminating. The revision emphasised harder reasoning questions and expanded the answer choices from four to ten. Its Hugging Face publication is recorded as May 2024.

How it was built

MMLU-Pro rebuilds the question collection by removing easier original items and using expanded sets of answer options. It covers fourteen academic and professional subjects in a multiple-choice format. The catalogue imports 200 test questions, a capped selection that should not be treated as the size of the complete benchmark.

At a glance

What a problem looks like

Multiple-choice questions with ten options. The import accepts question, options, category and src.

Read an actual problem

How it is scored

The source ships an answer with every problem; this catalogue keeps it in restricted storage and never renders it.

Reported scores

Evaluation conditions differ between rows, so this is not a ranking. Rows marked vendor are reports by the model’s own organization.

ModelScoreMetricSetupReported bySource
DeepSeek-V3DeepSeek64.4Acc.Base model; 5-shot.vendor (self-reported)Base Model, Standard Benchmarks, MMLU-Pro (Acc.), final column

Why it is hard

Ten options rather than four, and the easiest original items removed, so guessing and memorisation carry much less of the score.

Limitations and criticism

Ten options reduce the probability of uniform random guessing but do not prevent memorisation or exploitation of answer cues. A correct choice does not directly validate a model's explanation or its ability to solve an open-ended version of the task. Subject composition, prompting and answer extraction remain relevant to comparisons, and reuse of public question material leaves a risk of training exposure.

How much room is left

The recorded vendor report gives DeepSeek-V3 base 64.4 percent accuracy with five examples in the prompt. That result remains well below the ceiling for the reported setup and supports treating the test as challenging for that model. It does not establish the current leading score or the remaining headroom for every model family.

Discussion

1 thread and 0 posts on this page. Anyone may reply; no account is required.

Contribute

Share a reproduction, report an evaluation issue, or suggest a correction to this benchmark overview.

Tasks

Whole public benchmark.

Public problems
12,032
Formats
text (12,032)
Problems with images
0
Answer availability
Answer published by the source (12,032)
Configs
default (12,032)
Splits
test (12,032)
Awaiting a first discussion
12,032
Problems with a discussion
0
Public contributions
0

Provenance and terms

Pinned source revisions, newest first:

  • b189ec765aa7ed75c8acfea42df31fdae71f97be · acquired 2026-09-11 · release f17f6fbd935b
  • b189ec765aa7ed75c8acfea42df31fdae71f97be · acquired 2026-09-11 · release 518b10e2176c
  • b189ec765aa7ed75c8acfea42df31fdae71f97be · acquired 2026-09-11 · release c2f53d6ad184