language benchmark
Global-MMLU-Lite Leaderboard
Global-MMLU-Lite is a lightweight version of the Global MMLU benchmark for evaluating language models across multiple languages while addressing cultural and linguistic biases in multilingual evaluation.
Updated Sep 5, 2026
Models15
Model coverage15
MetricScore
EvidenceB
Global-MMLU-Lite Ranking
Higher score ranks better on this benchmark.
Sort by
Rank | Model | Score | Percentile | Participants | Evidence | Evaluated |
|---|
| Rank01 | ModelGO | Score89.20% | Percentile100.00% | Participants15 | EvidenceC | Evaluated |
| Rank02 | ModelGO | Score88.60% | Percentile92.86% | Participants15 | EvidenceC | Evaluated |
| Rank03 | ModelGO | Score88.40% | Percentile85.71% | Participants15 | EvidenceC | Evaluated |
| Rank04 | ModelTM | Score86.70% | Percentile78.57% | Participants15 | EvidenceC | Evaluated |
| Rank05 | ModelGO | Score81.10% | Percentile71.43% | Participants15 | EvidenceC | Evaluated |
| Rank06 | ModelGO | Score78.20% | Percentile64.29% | Participants15 | EvidenceC | Evaluated |
| Rank07 | ModelGO | Score75.10% | Percentile57.14% | Participants15 | EvidenceC | Evaluated |
| Rank08 | ModelGO | Score69.50% | Percentile50.00% | Participants15 | EvidenceC | Evaluated |
| Rank09 | ModelGO | Score69.10% | Percentile42.86% | Participants15 | EvidenceC | Evaluated |
| Rank10 | ModelGO | Score64.50% | Percentile35.71% | Participants15 | EvidenceC | Evaluated |
| Rank11 | ModelGO | Score64.50% | Percentile28.57% | Participants15 | EvidenceC | Evaluated |
| Rank12 | ModelGO | Score59.00% | Percentile21.43% | Participants15 | EvidenceC | Evaluated |
| Rank13 | ModelGO | Score59.00% | Percentile14.29% | Participants15 | EvidenceC | Evaluated |
| Rank14 | ModelGO | Score54.50% | Percentile7.14% | Participants15 | EvidenceC | Evaluated |
| Rank15 | ModelGO | Score34.20% | Percentile0.00% | Participants15 | EvidenceC | Evaluated |
Global-MMLU-Lite Highlights
The leading models and scores on this benchmark.
Global-MMLU-Lite Score Distribution
A closer view of the leading scores on this benchmark.
The Top AI Models for Global-MMLU-Lite
The first five results on this benchmark, with official price and output speed added where the model identity can be matched.
What is Global-MMLU-Lite?
What Global-MMLU-Lite measures and how its scores work.
Global-MMLU-Lite is a language benchmark and a lightweight version of the Global MMLU benchmark.
It measures language-model performance across multiple languages using the Score metric, reported as a ratio.
Scores are shown in ratio. This benchmark is not independently verified and has an evidence level of B.
- Family
- Global-MMLU-Lite
- Modality
- text
- Primary category
- language
- Score direction
- higher
- LLMBoard eligible
- No
- Evaluation key
- overall
LLM Stats. Benchmark scores retain their original unit. Overall score eligibility is shown separately.
FAQ
Common questions about Global-MMLU-Lite.
Which model scores highest on Global-MMLU-Lite?
Gemini 2.5 Pro Preview 06-05 is currently ranked first with 89.20%.
What are the top three models on Global-MMLU-Lite?
The current leaders are Gemini 2.5 Pro Preview 06-05 (89.20%), Gemini 2.5 Pro (88.60%), and Gemini 2.5 Flash (88.40%).
Which Global-MMLU-Lite model has the lowest official input price?
Gemini 2.5 Flash-Lite has the lowest matched official input price at $0.10 input / $0.40 output per 1M tokens.
Which models are fastest among Global-MMLU-Lite results?
The fastest matched records are Gemini 2.5 Pro Preview 06-05 (85.00 tok/s via Google), Gemini 2.5 Flash (85.00 tok/s via Google), and Gemini 2.0 Flash-Lite (85.00 tok/s via Google).
Does the highest score result prove overall model quality?
No. This benchmark measures one defined capability or task. The overall LLMBoard score uses a separate aggregation across eligible benchmark evidence.