HAE-RAE Math 8K (HRM8K)
We show this table for reference; we do not rank on it.
Korean mathematical reasoning (high-school to Olympiad level).
Benchmark score on HRM8K — November 13, 2026
We compile the HRM8K rows from provider self-reports. Solar Open 2 leads the table at 92.2%, followed by K-Exaone (90.9%). We do not use these results to rank models overall.
Solar Open 2
Upstage
Reported by UpstageK-Exaone
LG AI Research
Reported by LG AI Research2 modelsKoreanKorean-language benchmarkRefreshingDisplay onlyUpdated November 13, 2026
Benchmark score table (2 models)
All 2 results on this page were reported by the labs that built the models.
| Rank | Model / configuration | Score | Who ran it | Parameters (B) | Open / closed |
|---|---|---|---|---|---|
| 1 | Solar Open 2UpstageReported by Upstage | 92.2% | Reported by Upstage | Not reported | Open |
| 2 | K-ExaoneLG AI ResearchReported by LG AI Research | 90.9% | Reported by LG AI Research | Not reported | Closed |
- Lab-reported
- Benchmark-run
- Independent
About HRM8K
Tasks
8,011 instances
Format
Math word problems
Difficulty
Olympiad level
Freshness and provenance
Version
HRM8K
Refresh cadence
Static
Staleness state
Refreshing
Question availability
Public benchmark set
BenchLM uses freshness metadata to decide whether a benchmark should still be treated as a strong differentiator, a benchmark to watch, or a display-only reference. For the full scoring policy, see the BenchLM methodology page.
Questions
What does HRM8K measure?
Korean mathematical reasoning (high-school to Olympiad level).
Which model scores highest on HRM8K?
Solar Open 2 by Upstage currently leads with a score of 92.2% on HRM8K.
How many models are evaluated on HRM8K?
2 AI models have published results on HRM8K in the BenchLM catalog.
Compare top models on HRM8K
Know when it’s worth switching models
The model to choose, the cheaper alternative, and the release we would wait on.
Read a sample issueJoin 7,000+ subscribers.
One email each week. Unsubscribe anytime.