| Benchmark 分类 | Benchmark | Temperature | Max token 推荐设置 | 推荐测试次数 | Top-P | 其他 |
|---|---|---|---|---|---|---|
| Multi-modal | MMMU-Pro | 推荐设置:1.0 | max tokens = 96k | 3次 | top\_p=0.95 | thinking= |
| MMMU-Pro w/ python | 推荐设置:1.0 | per step tokens = 64k; total max tokens = 256k |
3次 | top\_p=0.95 | 推荐max steps = 50 thinking= |
|
| CharXiv (RQ) | 推荐设置:1.0 | max tokens = 96k | 3次 | top\_p=0.95 | thinking= | |
| CharXiv (RQ) w/ python | 推荐设置:1.0 | per step tokens = 64k; total max tokens = 256k |
3次 | top\_p=0.95 | 推荐max steps = 50 thinking= |
|
| MathVision | 推荐设置:1.0 | max tokens = 96k | 3次 | top\_p=0.95 | thinking= | |
| MathVision w/ python | 推荐设置:1.0 | per step tokens = 64k; total max tokens = 256k |
3次 | top\_p=0.95 | 推荐max steps = 50 thinking= |
|
| V\* w/ python | 推荐设置:1.0 | per step tokens = 64k; total max tokens = 256k |
3次 | top\_p=0.95 | 推荐max steps = 50 thinking= |
|
| Agent | HLE-Full w/ tools | 推荐设置:1.0 | per step tokens = 48k; total max tokens = 256k |
1次 | top\_p=0.95 | 推荐max steps = 300 thinking= |
| BrowseComp | 推荐设置:1.0 | per step tokens = 48k; total max tokens = 256k |
1次 | top\_p=0.95 | 推荐max steps = 300 thinking= |
|
| DeepSearchQA | 推荐设置:1.0 | per step tokens = 48k; total max tokens = 256k |
1次 | top\_p=0.95 | 推荐max steps = 300 thinking= |
|
| WideSearch | 推荐设置:1.0 | per step tokens = 48k; total max tokens = 256k |
4次 | top\_p=0.95 | 推荐max steps = 300 thinking= |
|
| Toolathlon | 推荐设置:1.0 | per step tokens = 48k; total max tokens = 256k |
4次 | top\_p=0.95 | 推荐max steps = 300 thinking= |
|
| MCPMark | 推荐设置:1.0 | per step tokens = 48k; total max tokens = 256k |
4次 | top\_p=0.95 | 推荐max steps = 300 thinking= |
|
| Claw Eval | 推荐设置:1.0 | per step tokens = 48k; total max tokens = 256k |
4次 | top\_p=0.95 | 推荐max steps = 300 thinking= |
|
| APEX-Agents | 推荐设置:1.0 | per step tokens = 48k; total max tokens = 256k |
4次 | top\_p=0.95 | 推荐max steps = 300 thinking= |
|
| Coding | Terminal-Bench 2.0 (Terminus-2) | 推荐设置:1.0 | max tokens = 256k | 3次 | top\_p=0.95 | thinking= |
| SWE-Bench Pro | 推荐设置:1.0 | per step tokens = 32k; total max tokens = 256k |
5次 | top\_p=0.95 | 推荐max steps = 300 thinking= |
|
| SWE-Bench Multilingual | 推荐设置:1.0 | per step tokens = 32k; total max tokens = 256k |
5次 | top\_p=0.95 | 推荐max steps = 300 thinking= |
|
| SWE-Bench Verified | 推荐设置:1.0 | per step tokens = 32k; total max tokens = 256k |
5次 | top\_p=0.95 | 推荐max steps = 300 thinking= |
|
| SciCode | 推荐设置:1.0 | max tokens = 96k | 4次 | top\_p=0.95 | thinking= | |
| OJBench (python) | 推荐设置:1.0 | max tokens = 96k | 8次 | top\_p=0.95 | thinking= | |
| LiveCodeBench (v6) | 推荐设置:1.0 | max tokens = 96k | 1次 | top\_p=0.95 | thinking= | |
| Math | AIME 2026 | 推荐设置:1.0 | max tokens = 96k | 32次 | top\_p=0.95 | thinking= |
| HMMT 2026 (Feb) | 推荐设置:1.0 | max tokens = 96k | 32次 | top\_p=0.95 | thinking= | |
| IMO-AnswerBench | 推荐设置:1.0 | max tokens = 96k | 4次 | top\_p=0.95 | thinking= | |
| Knowledge | HLE-Full | 推荐设置:1.0 | max tokens = 96k | 1次 | top\_p=0.95 | thinking= |
| GPQA-Diamond | 推荐设置:1.0 | max tokens = 96k | 8次 | top\_p=0.95 | thinking= |