MathArena Apex 2025
MathArena Apex 2025 - hardest math contest problems of the 2025 MathArena set.
Category
STEM & Reasoning
Score type
percent
Scale
100%
higher is better
Models tested
5
What is MathArena Apex 2025?
MathArena Apex 2025 is a benchmark for hard reasoning in science, technology, engineering and math. The official description says: "MathArena Apex 2025 - hardest math contest problems of the 2025 MathArena set". Below we describe in simple words what benchmarks of this type generally examine.
What does it test?
MathArena Apex 2025 measures hard reasoning in science, technology, engineering and math. In detail, it checks:
- graduate- and expert-level questions in science and math
- multi-step derivations where each step must be correct
- problems written to defeat memorisation (novel or recent exam problems)
- correct use of formulas, units and logical inference
How to read the score
The score is accuracy: the percentage of tasks answered or solved correctly. 100% means every task was solved perfectly; higher is better.
Official methodology note: MathArena Apex 2025 - hardest math contest problems of the 2025 MathArena set.
Frequently asked questions
What is the MathArena Apex 2025 benchmark?
MathArena Apex 2025 is a benchmark for hard reasoning in science, technology, engineering and math. The official description says: 'MathArena Apex 2025 - hardest math contest problems of the 2025 MathArena set'. Below we describe in simple words what benchmarks of this type generally examine.
What does MathArena Apex 2025 test?
MathArena Apex 2025 mainly measures hard reasoning in science, technology, engineering and math. Concretely, it checks: graduate- and expert-level questions in science and math; multi-step derivations where each step must be correct; problems written to defeat memorisation (novel or recent exam problems).
How is the MathArena Apex 2025 score calculated?
The score is accuracy: the percentage of tasks answered or solved correctly. 100% means every task was solved perfectly; higher is better. The declared maximum score of this benchmark is 100.0.
Where do I get MathArena Apex 2025 data and scores?
The official benchmark page/reference is linked on this page. Every score on Open LLM Radar is also available as open data: download the scores JSON (http://open-llm-radar-api:8000/api/v1/benchmarks/matharena-apex-2025/scores) or pull it directly from the Open LLM Radar API with any HTTP client.
Benchmark data & downloads
Where to get, inspect and reuse the data behind this benchmark.
Benchmark dataset
The question/task data itself — download and evaluate locally
Benchmark metadata (JSON)
This page's benchmark record from the Open LLM Radar API
All model scores (JSON download)
Full score table as JSON; paginate with limit/offset params
Open LLM Radar API docs
Swagger UI to explore and reuse all radar API endpoints
Model scores 5 models tested
| # | Model | Score | Date | Source | Verified |
|---|---|---|---|---|---|
| 1 |
Hy4 preview
Tencent
|
74.2%
|
28.08.2026 | Tencent Hy4 preview model card benchmark appendix (https://huggingface.co/tencent/Hy4-preview#benchmark-appendix) | Verified |
| 2 |
Qwen3.8-2.4T-A95B
Qwen
|
72.8%
|
28.08.2026 | Tencent Hy4 preview model card benchmark appendix (https://huggingface.co/tencent/Hy4-preview#benchmark-appendix) (Tencent rerun) | Verified |
| 3 |
Kimi K3
Moonshot AI
|
68.4%
|
28.08.2026 | Tencent Hy4 preview model card benchmark appendix (https://huggingface.co/tencent/Hy4-preview#benchmark-appendix) (Tencent rerun) | Verified |
| 4 |
DeepSeek-V4.1-Flash
DeepSeek
|
65.6%
|
10.09.2026 | Report | Verified |
| 5 |
Hy3
Tencent
|
38.7%
|
28.08.2026 | Tencent Hy4 preview model card benchmark appendix (https://huggingface.co/tencent/Hy4-preview#benchmark-appendix) | Verified |