Skip to content

Benchmark

LaoBench Lao-7k K12 Education Natural Sciences accuracy (zero-shot)

Multilingual benchmarks compare model performance across languages. Scores are percentages. Higher is better.

Models and systems measured
14
Scores recorded
15
Sources
1
IQ weight
1.20

Score distribution

78.9%
Mean
86%
Median
19.8%
Std dev
35.2%
10th percentile
90%
90th percentile
25%
Min
25%98.3%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.20 for chat models, measured over 10 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#Model or systemScore
1GPT 5table: 3, mode: single· LaoBench (arXiv:2511.11334v3) data90%
2Claude Sonnet 4.5table: 3, mode: single· LaoBench (arXiv:2511.11334v3) data89.4%
3Claude Opus 4.1table: 3, mode: single· LaoBench (arXiv:2511.11334v3) data89%
4Gemini 2.5 Protable: 3, mode: single· LaoBench (arXiv:2511.11334v3) data88.7%
5Qwen3 Maxtable: 3, mode: single· LaoBench (arXiv:2511.11334v3) data87.3%
6Qwen3 235B A22B Instruct 2507table: 3, mode: single· LaoBench (arXiv:2511.11334v3) data86.5%
7DeepSeek V3.2 Exptable: 3, mode: Thinking· LaoBench (arXiv:2511.11334v3) data84.7%
8Qwen 3 235B A22Btable: 3, mode: single· LaoBench (arXiv:2511.11334v3) data82.8%
9DeepSeek V3.2 Exptable: 3, mode: Non-Thinking· LaoBench (arXiv:2511.11334v3) data80%
10Qwen3 Next 80B A3B Instructtable: 3, mode: single· LaoBench (arXiv:2511.11334v3) data76.2%
11Ling Mini 2.0table: 3, mode: single· LaoBench (arXiv:2511.11334v3) data35.2%

Not ranked

The source marks these results as descriptive, not capability scores, so we list them by name instead of ranking them.

Model or systemScore
Human Expertstable: 3, mode: single· LaoBench (arXiv:2511.11334v3) data98.3%
Qwen3-PlusNon-Thinkingtable: 3, mode: Non-Thinking· LaoBench (arXiv:2511.11334v3) data86%
Qwen3-PlusThinkingtable: 3, mode: Thinking· LaoBench (arXiv:2511.11334v3) data84.9%
Random Choicetable: 3, mode: single· LaoBench (arXiv:2511.11334v3) data25%

Data credits: LaoBench, BAAI et al., arXiv:2511.11334v3. Cited numeric results only. arXiv non-exclusive distribution notice establishes no results reuse grant; dataset Apache licence is not a results grant.