UnifyBench

RankingMMLU-Pro

MMLU-Pro

Data updated 22 Oct 2026

Bucket
Knowledge
Unit
percent
Direction
Higher is better
Version
—
Display harness
MMLU-Pro reported
Board
https://github.com/TIGER-AI-Lab/MMLU-Pro

Compare published benchmark results with category weights →

Models

Results are ordered by score in the display harness, followed by models with no result. Other harnesses are listed separately below. Missing results remain unknown.

26–50 of 471 entries

ModelScoreEvidenceSource-recorded date
Phi-4-reasoning-plusMicrosoft76%official board2026-10-13
MiniMax-Text-01MiniMax75.7%official board2026-10-13
Llama-4-Scout-17B-16E-InstructMeta74.3%official board2026-10-13
Phi-4-reasoningMicrosoft74.3%official board2026-10-13
Llama-3.1-405B-InstructMeta73.3%official board2026-10-13
phi-4Microsoft70.4%official board2026-10-13
Qwen3-235B-A22BQwen68.18%official board2026-10-13
Gemma 3 27B ITGoogle67.5%official board2026-10-13
Llama-3.3-70B-InstructMeta65.92%official board2026-10-13
GPT-4-turboOpenAI63.71%official board2026-10-13
GPT-4o MiniOpenAI63.09%official board2026-10-13
Llama-3.1-70B-InstructMeta62.84%official board2026-10-13
Gemma 3 12B ITGoogle60.6%official board2026-10-13
EXAONE 3.5 32B InstructLG AI Research58.91%official board2026-10-13
MiMo-7B-RLXiaomi58.6%official board2026-10-13
Gemma 2 27B ITGoogle56.54%official board2026-10-13
Qwen3.5-2BQwen55.3%official board2026-10-13
Gemma 2 9B ITGoogle52.08%official board2026-10-13
glm-4-9b-chatZ.ai48.01%official board2026-10-13
Phi-3.5-mini-instructMicrosoft47.87%official board2026-10-13
EXAONE 3.5 7.8B InstructLG AI Research46.24%official board2026-10-13
Llama-3.1-8B-InstructMeta44.25%official board2026-10-13
Gemma 3 4B ITGoogle43.6%official board2026-10-13
EXAONE 3.5 2.4B InstructLG AI Research39.1%official board2026-10-13
aya-expanse-8bCohere33.74%official board2026-10-13