Benchmark
ArXivMath Benchmark
What is ArXivMath?
Competition-style mathematics drawn from arXiv; test-time compute budget is part of the result. Benchmark Radar tracks 1 reported scores for ArXivMath, cited…
- Evaluates
- math
- Openness
- unknown
- Importer
- Model reports
- Reported scores
- 1
Source provenance
- Original evidence https://matharena.ai/arxivmath
ArXivMath links
- Website https://matharena.ai/arxivmath
ArXivMath results and reported scores
Model reports
| Model | Organization | Reported value | Reported | Evidence |
|---|---|---|---|---|
| Hy4 preview | Tencent | 66.6 | 2026-08-28 | evidence |
Scores are partitioned by the source that reported them and are never merged into a single cross-source ranking, because the sources measure different things and say so.
Which sources cite ArXivMath?
- Hy4 preview Tencent, 2026-08-28