Mathematical Reasoning on AMO-Bench (Avg@5)
0.646Avg@5GPT-5.1-high
Evaluation results
| Method | Links | |
|---|---|---|
| GPT-5.1-highEvaluation Variant=VeRA-H, Sample Count=2442026.01 | 0.646 | |
| Gemini-3-Pro-PreviewEvaluation Variant=VeRA-H, Sample Count=2442026.01 | 0.611 | |
| GPT-5.1-highEvaluation Variant=VeRA-H Pro, Sample Count=502026.01 | 0.588 | |
| GPT-5-highEvaluation Variant=VeRA-H, Sample Count=2442026.01 | 0.584 | |
| GPT-5.1-highEvaluation Variant=seeds, Sample Count=502026.01 | 0.56 | |
| Gemini-3-Pro-PreviewEvaluation Variant=seeds, Sample Count=502026.01 | 0.56 | |
| Gemini-3-Pro-PreviewEvaluation Variant=VeRA-H Pro, Sample Count=502026.01 | 0.56 | |
| GPT-5-highEvaluation Variant=VeRA-H Pro, Sample Count=502026.01 | 0.544 | |
| GLM-4.6Evaluation Variant=VeRA-H, Sample Count=2442026.01 | 0.529 | |
| DeepSeek-V3.1-thinkingEvaluation Variant=VeRA-H Pro, Sample Count=502026.01 | 0.528 | |
| DeepSeek-V3.1-thinkingEvaluation Variant=VeRA-H, Sample Count=2442026.01 | 0.521 | |
| DeepSeek-V3.2-thinkingEvaluation Variant=VeRA-H, Sample Count=2442026.01 | 0.516 | |
| GLM-4.6Evaluation Variant=VeRA-H Pro, Sample Count=502026.01 | 0.48 | |
| Seed-1.6-1015-highEvaluation Variant=seeds, Sample Count=502026.01 | 0.48 | |
| DeepSeek-V3.1-thinkingEvaluation Variant=seeds, Sample Count=502026.01 | 0.48 | |
| DeepSeek-V3.2-thinkingEvaluation Variant=VeRA-H Pro, Sample Count=502026.01 | 0.46 | |
| Gemini-2.5-ProEvaluation Variant=VeRA-H, Sample Count=2442026.01 | 0.435 | |
| Seed-1.6-1015-highEvaluation Variant=VeRA-H, Sample Count=2442026.01 | 0.431 | |
| Seed-1.6-Thinking-0715Evaluation Variant=VeRA-H, Sample Count=2442026.01 | 0.416 | |
| GLM-4.6Evaluation Variant=seeds, Sample Count=502026.01 | 0.4 | |
| Seed-1.6-Thinking-0715Evaluation Variant=seeds, Sample Count=502026.01 | 0.4 | |
| GPT-5-highEvaluation Variant=seeds, Sample Count=502026.01 | 0.4 | |
| Seed-1.6-Lite-1015-highEvaluation Variant=VeRA-H, Sample Count=2442026.01 | 0.397 | |
| qwen3-max-0923Evaluation Variant=VeRA-H, Sample Count=2442026.01 | 0.392 | |
| Seed-1.6-Thinking-0715Evaluation Variant=VeRA-H Pro, Sample Count=502026.01 | 0.38 | |
| Gemini-2.5-ProEvaluation Variant=VeRA-H Pro, Sample Count=502026.01 | 0.376 | |
| Minimax-M2Evaluation Variant=VeRA-H, Sample Count=2442026.01 | 0.366 | |
| Seed-1.6-1015-highEvaluation Variant=VeRA-H Pro, Sample Count=502026.01 | 0.364 | |
| Seed-1.6-Lite-1015-highEvaluation Variant=VeRA-H Pro, Sample Count=502026.01 | 0.364 | |
| Claude-Sonnet-4.5-thinkingEvaluation Variant=VeRA-H, Sample Count=2442026.01 | 0.36 | |
| Seed-1.6-Lite-1015-highEvaluation Variant=seeds, Sample Count=502026.01 | 0.36 | |
| qwen3-max-0923Evaluation Variant=VeRA-H Pro, Sample Count=502026.01 | 0.352 | |
| Claude-Sonnet-4.5-thinkingEvaluation Variant=VeRA-H Pro, Sample Count=502026.01 | 0.324 | |
| Minimax-M2Evaluation Variant=VeRA-H Pro, Sample Count=502026.01 | 0.284 | |
| Gemini-2.5-ProEvaluation Variant=seeds, Sample Count=502026.01 | 0.28 | |
| DeepSeek-V3.2-thinkingEvaluation Variant=seeds, Sample Count=502026.01 | 0.28 | |
| Kimi-K2-thinkingEvaluation Variant=VeRA-H, Sample Count=2442026.01 | 0.265 | |
| Kimi-K2-thinkingEvaluation Variant=seeds, Sample Count=502026.01 | 0.22 | |
| GPT-5.1-chat-latestEvaluation Variant=VeRA-H, Sample Count=2442026.01 | 0.207 | |
| Minimax-M2Evaluation Variant=seeds, Sample Count=502026.01 | 0.2 | |
| Kimi-K2-0905Evaluation Variant=VeRA-H, Sample Count=2442026.01 | 0.199 | |
| Kimi-K2-0905Evaluation Variant=VeRA-H Pro, Sample Count=502026.01 | 0.188 | |
| Claude-Sonnet-4.5-thinkingEvaluation Variant=seeds, Sample Count=502026.01 | 0.18 | |
| GPT-5.1-chat-latestEvaluation Variant=VeRA-H Pro, Sample Count=502026.01 | 0.176 | |
| Kimi-K2-thinkingEvaluation Variant=VeRA-H Pro, Sample Count=502026.01 | 0.172 | |
| qwen3-max-0923Evaluation Variant=seeds, Sample Count=502026.01 | 0.14 | |
| Kimi-K2-0905Evaluation Variant=seeds, Sample Count=502026.01 | 0.08 | |
| GPT-5.1-chat-latestEvaluation Variant=seeds, Sample Count=502026.01 | 0.06 |