HorizonMath (pass@4)
HorizonMath - long-horizon math reasoning, pass@4 protocol, reported in the Tencent Hy4 preview model card appendix.
Category
STEM & Reasoning
Max Score
100.0
Score Type
percent
Active
Yes
Model Scores
Scores for HorizonMath (pass@4)
| Model | Score | Date | Verified |
|---|---|---|---|
| Hy3 |
3.50
|
28.08.2026 | Verified |
| Hy4 preview |
74.44%
|
28.08.2026 | Verified |
| DeepSeek V4 Pro |
12.92%
|
28.08.2026 | Verified |
| Qwen3.8-2.4T-A95B |
25.42%
|
28.08.2026 | Verified |
| Kimi K3 |
50.28%
|
28.08.2026 | Verified |
| GPT-5.6 Sol |
100.00%
|
28.08.2026 | Verified |
| Claude Opus 5 |
25.28%
|
28.08.2026 | Verified |