Benchmark
MathArena Apex
MathArena Apex is a challenging math contest benchmark featuring the most difficult mathematical problems designed to test advanced reasoning and…
- Modality
- text
- Categories
- math, reasoning
- Openness
- unknown
- Source
- llm_stats
- Reported scores
- 8
Reported scores
llm_stats
| Model | Organization | Reported value | Reported | Evidence |
|---|---|---|---|---|
| DeepSeek-V4-Flash-0423 | DeepSeek | 0.721 | 2026-04-23 | evidence |
| DeepSeek-V4-Flash-Max | DeepSeek | 0.857 | 2026-04-23 | evidence |
| DeepSeek-V4-Pro-Max | DeepSeek | 0.902 | 2026-04-23 | evidence |
| Gemini 3 Pro | 0.234 | 2025-11-18 | evidence | |
| Hy3 | Tencent | 0.387 | 2026-07-06 | evidence |
| Qwen3.7 Max | Qwen | 0.445 | 2026-05-19 | evidence |
| Seed 2.1 Pro | ByteDance | 0.313 | 2026-06-24 | evidence |
| Seed 2.1 Turbo | ByteDance | 0.354 | 2026-06-24 | 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.