| GDPval-AA (Agnetic Real-World Tasks, (ELO-500)/2000) | OpenAI | 100 % | 估读 |
| GDPval-AA (Agnetic Real-World Tasks, (ELO-500)/2000) | Anthropic | 100 % | 估读 |
| GDPval-AA (Agnetic Real-World Tasks, (ELO-500)/2000) | Google | 100 % | 估读 |
| GDPval-AA (Agnetic Real-World Tasks, (ELO-500)/2000) | Alibaba | 100 % | 估读 |
| GDPval-AA (Agnetic Real-World Tasks, (ELO-500)/2000) | xAI | 100 % | 估读 |
| GDPval-AA (Agnetic Real-World Tasks, (ELO-500)/2000) | Meta | 100 % | 估读 |
| GDPval-AA (Agnetic Real-World Tasks, (ELO-500)/2000) | DeepSeek | 100 % | 估读 |
| GDPval-AA (Agnetic Real-World Tasks, (ELO-500)/2000) | Mistral | 100 % | 估读 |
| GDPval-AA (Agnetic Real-World Tasks, (ELO-500)/2000) | NVIDIA | 100 % | 估读 |
| GDPval-AA (Agnetic Real-World Tasks, (ELO-500)/2000) | Amazon | 100 % | 估读 |
| GDPval-AA (Agnetic Real-World Tasks, (ELO-500)/2000) | Zhipu AI | 100 % | 估读 |
| GDPval-AA (Agnetic Real-World Tasks, (ELO-500)/2000) | Moonshot AI | 100 % | 估读 |