MMLU-Redux
An improved version of the MMLU benchmark featuring manually re-annotated questions to identify and correct errors in the original dataset. Provides more reliable evaluation metrics for language models by addressing dataset quality issues found in the original MMLU.
Qwen3.5-397B-A17B from Alibaba Cloud / Qwen Team currently leads the MMLU-Redux leaderboard with a score of 0.949 across 45 evaluated AI models.
Qwen3.5-397B-A17B leads with 94.9%, followed by Qwen3.6 Plus at 94.5% and 
Kimi K2-Thinking-0905 at 94.4%.
Progress Over Time
Interactive timeline showing model performance evolution on MMLU-Redux
MMLU-Redux Leaderboard
| Context | Cost | License | ||||
|---|---|---|---|---|---|---|
| 1 |  Alibaba Cloud / Qwen Team | 397B | 262K | $0.60 / $3.60 | ||
| 2 | Alibaba Cloud / Qwen Team | — | 1.0M | $0.50 / $3.00 | ||
| 3 | Moonshot AI | 1.0T | — | — | ||
| 4 | Alibaba Cloud / Qwen Team | 122B | 262K | $0.40 / $3.20 | ||
| 5 | Alibaba Cloud / Qwen Team | 235B | 262K | $0.30 / $3.00 | ||
| 6 | Alibaba Cloud / Qwen Team | 236B | 262K | $0.45 / $3.49 | ||
| 7 | Alibaba Cloud / Qwen Team | 28B | 262K | $0.60 / $3.60 | ||
| 8 |  DeepSeek | 671B | 131K | $0.55 / $2.19 | ||
| 9 | Alibaba Cloud / Qwen Team | 35B | — | — | ||
| 9 | Alibaba Cloud / Qwen Team | 35B | 262K | $0.25 / $2.00 | ||
| 11 | Alibaba Cloud / Qwen Team | 27B | 262K | $0.30 / $2.40 | ||
| 12 | Alibaba Cloud / Qwen Team | 235B | 262K | $0.15 / $0.80 | ||
| 13 | Moonshot AI | 1.0T | — | — | ||
| 13 | Moonshot AI | 1.0T | 200K | $0.50 / $0.50 | ||
| 15 | Alibaba Cloud / Qwen Team | 80B | 66K | $0.15 / $1.50 | ||
| 16 | Alibaba Cloud / Qwen Team | 236B | 262K | $0.30 / $1.49 | ||
| 17 | Alibaba Cloud / Qwen Team | 33B | — | — | ||
| 18 | DeepSeek | 671B | 164K | $0.27 / $1.00 | ||
| 19 | Alibaba Cloud / Qwen Team | 9B | — | — | ||
| 20 | Alibaba Cloud / Qwen Team | 80B | 66K | $0.15 / $1.50 | ||
| 20 | Alibaba Cloud / Qwen Team | 31B | 262K | $0.20 / $1.00 | ||
| 22 | Alibaba Cloud / Qwen Team | 33B | — | — | ||
| 23 |  Meituan | 560B | — | — | ||
| 24 | DeepSeek | 671B | 131K | $0.27 / $1.10 | ||
| 25 | Alibaba Cloud / Qwen Team | 9B | 262K | $0.18 / $2.09 | ||
| 25 | Alibaba Cloud / Qwen Team | 4B | — | — | ||
| 27 | Alibaba Cloud / Qwen Team | 31B | 262K | $0.20 / $0.70 | ||
| 28 | Alibaba Cloud / Qwen Team | 235B | 128K | $0.10 / $0.10 | ||
| 29 | Alibaba Cloud / Qwen Team | 73B | 131K | $0.35 / $0.40 | ||
| 30 | Alibaba Cloud / Qwen Team | 4B | 262K | $0.10 / $1.00 | ||
| 31 | Alibaba Cloud / Qwen Team | 9B | 262K | $0.08 / $0.50 | ||
| 32 | Alibaba Cloud / Qwen Team | 33B | — | — | ||
| 33 |  Mistral AI | 14B | — | — | ||
| 33 | Mistral AI | 675B | 128K | $2.00 / $5.00 | ||
| 35 | Alibaba Cloud / Qwen Team | 4B | 262K | $0.10 / $0.60 | ||
| 36 | Alibaba Cloud / Qwen Team | 15B | — | — | ||
| 37 | Alibaba Cloud / Qwen Team | 2B | — | — | ||
| 38 | Mistral AI | 8B | — | — | ||
| 39 | Alibaba Cloud / Qwen Team | 32B | 128K | $0.09 / $0.09 | ||
| 40 | Alibaba Cloud / Qwen Team | 8B | 131K | $0.30 / $0.30 | ||
| 41 | Mistral AI | 3B | — | — | ||
| 42 | Alibaba Cloud / Qwen Team | 7B | — | — | ||
| 43 | Alibaba Cloud / Qwen Team | 7B | — | — | ||
| 44 | Alibaba Cloud / Qwen Team | 800M | — | — | ||
| 45 |  Baidu | 21B | 128K | $0.40 / $4.00 |
FAQ
Common questions about MMLU-Redux.
More evaluations to explore
Related benchmarks in the same category
A challenging dataset of 448 multiple-choice questions written by domain experts in biology, physics, and chemistry. Questions are Google-proof and extremely difficult, with PhD experts reaching 65% accuracy.
A more robust and challenging multi-task language understanding benchmark that extends MMLU by expanding multiple-choice options from 4 to 10, eliminating trivial questions, and focusing on reasoning-intensive tasks. Features over 12,000 curated questions across 14 domains and causes a 16-33% accuracy drop compared to original MMLU.
All 30 problems from the 2025 American Invitational Mathematics Examination (AIME I and AIME II), testing olympiad-level mathematical reasoning with integer answers from 000-999. Used as an AI benchmark to evaluate large language models' ability to solve complex mathematical problems requiring multi-step logical deductions and structured symbolic reasoning.
Massive Multitask Language Understanding benchmark testing knowledge across 57 diverse subjects including STEM, humanities, social sciences, and professional domains
A verified subset of 500 software engineering problems from real GitHub issues, validated by human annotators for evaluating language models' ability to resolve real-world coding issues by generating patches for Python codebases.
Humanity's Last Exam (HLE) is a multi-modal academic benchmark with 2,500 questions across mathematics, humanities, and natural sciences, designed to test LLM capabilities at the frontier of human knowledge with unambiguous, verifiable solutions