Global-MMLU-Lite (AA)
Knowledge questions across languages rather than only English. Run by Artificial Analysis; too few models so far to rank on.
As of 24 Oct 2026, Gemini 3.1 Pro leads Global-MMLU-Lite (AA) on BenchLeader with 93.2%, ahead of Gemini 3 Flash at 92.8%, across 11 model configurations with a published result.
- Published by
- Artificial Analysis
- Category
- Knowledge
- Index weight
- Reference only
- Models
- 11
- Data as of
- 24 Oct 2026
- Newest result
- 19 Feb 2026
The publisher has posted no new results on this board since 19 Feb 2026. BenchLeader re-reads it every day, so newer models appear here as soon as they are published.
Source: Artificial Analysis (artificialanalysis.ai). Data taken from the public leaderboard.
- 1Gemini 3.1 Pro93.2%
- 2Gemini 3 Flash (none)92.8%
- 3Gemini 3 Pro (high)92.2%
- 4Claude Opus 4.6 (max)92.2%
- 5Gemini 3 Flash (thinking)91.4%
- 6Intellect 380.3%
- 7K2 Think V2 (high)78.6%
- 8K2 Think V2 (medium)76.7%
- 9Olmo 3.1 32B (thinking)71.5%
- 10K2 Think V2 (low)71.4%
- 11Nemotron 3 Nano 30B A3B (none)61.2%
11 of 11
| # | ||||
|---|---|---|---|---|
| 1 |  Gemini 3.1 Pro | 93.2% | 60.6 | 2026-02-19 |
| 2 | Gemini 3 Flashno reasoning | 92.8% | 53.3 | 2025-12-17 |
| 3 | Gemini 3 Prohigh | 92.2% | 60.0 | 2025-11-18 |
| 4 |  Claude Opus 4.6max | 92.2% | 57.4 | 2026-02-05 |
| 5 | Gemini 3 Flashthinking | 91.4% | 59.1 | 2025-12-17 |
| 6 |  Intellect 3open ↗ | 80.3% | 46.5 | 2025-11-27 |
| 7 |  K2 Think V2highopen | 78.6% | 45.8 | 2025-12-05 |
| 8 | K2 Think V2mediumopen | 76.7% | 45.0 | 2025-12-05 |
| 9 |  Olmo 3.1 32Bthinkingopen ↗ | 71.5% | 44.0 | 2025-12-12 |
| 10 | K2 Think V2lowopen | 71.4% | 42.6 | 2025-12-05 |
| 11 |  Nemotron 3 Nano 30B A3Bno reasoningopen ↗ | 61.2% | 41.3 | 2025-12-15 |
More knowledge benchmarks · All LLM benchmarks
Cite as: BenchLeader, “Global-MMLU-Lite (AA) leaderboard”, data as of 24 Oct 2026, https://www.benchleader.com/benchmarks/aa_global_mmlu_lite.
Global-MMLU-Lite (AA): questions
- What does Global-MMLU-Lite (AA) measure?
- Knowledge questions across languages rather than only English. Run by Artificial Analysis; too few models so far to rank on. Scores are reported in percent of tasks solved; higher is better.
- Which AI model leads Global-MMLU-Lite (AA)?
- Gemini 3.1 Pro leads Global-MMLU-Lite (AA) with 93.2% as of 24 Oct 2026, ahead of Gemini 3 Flash at 92.8%.
- How many models have Global-MMLU-Lite (AA) results?
- 11 model configurations have a Global-MMLU-Lite (AA) result on BenchLeader, all taken from Artificial Analysis.
- Who runs Global-MMLU-Lite (AA) and how often is it updated?
- Global-MMLU-Lite (AA) is published by Artificial Analysis. BenchLeader re-reads the published results once a day and records the date each result was published.
- Does Global-MMLU-Lite (AA) count toward the BenchLeader Index?
- No. Global-MMLU-Lite (AA) is shown for reference but left out of the composite index.