Benchmark
Beyond AIME
Beyond AIME is a difficult mathematical reasoning benchmark designed to test deeper reasoning chains and harder decomposition than standard AIME-style…
- Modality
- text
- Categories
- math, reasoning
- Openness
- unknown
- Source
- llm_stats
- Reported scores
- 5
Reported scores
llm_stats
| Model | Organization | Reported value | Reported | Evidence |
|---|---|---|---|---|
| Mistral Medium 3.5 | Mistral | 0.669 | 2026-04-29 | evidence |
| Sarvam-105B | Sarvam AI | 0.691 | 2026-03-06 | evidence |
| Sarvam-30B | Sarvam AI | 0.583 | 2026-03-06 | evidence |
| Seed 2.1 Pro | ByteDance | 0.87 | 2026-06-24 | evidence |
| Seed 2.1 Turbo | ByteDance | 0.88 | 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.