Multilingual benchmark
PolyMath leaderboard
Every model the catalog carries a published PolyMath value for, ranked by that value.
CategoryMultilingual
MeasureCross-lingual mathematical reasoning
TasksMultilingual math problems
DifficultyAdvanced multilingual reasoning
Published byQwen3.6 launch benchmarks
A multilingual mathematical reasoning benchmark that tests whether math performance transfers across languages rather than only in English.
PolyMath ranking
3 models with a published PolyMath value, ordered by that value, highest first. Models the source has not scored on this benchmark are not listed — they are unmeasured, not last.
| Rank | Model | Provider | Cross-lingual mathematical reasoning |
|---|---|---|---|
| 1 |  Qwen3.7 Max | Alibaba | 86.5 |
| 2 | Qwen3.7 Plus | Alibaba | 84.0 |
| 3 |  K-EXAONE 2.0 | LG AI Research | 71.3 |
Evidence key: Observed
Rows are ordered by the value Qwen3.6 launch benchmarks published, highest first. The source does not state whether a higher value is the better result, so this page does not either: for a benchmark that measures a rate of failure, read the table from the bottom.