- GPT0%
- Token+2.38%
动察 Beating AI 快讯,Artificial Analysis 更新 Intelligence Index v4.2,旨在让榜单更难被针对性刷分。私有测试权重从 20% 提高到 40%,模型厂商无法提前看到完整测试集,减少针对 Benchmark 专门优化的空间。
新版加入 AA-Briefcase 和 GDP.pdf。AA-Briefcase 模拟持续数周的真实知识工作项目,模型要处理大量关联任务和数千份文件。GDP.pdf 则拿 100 份、共 4592 页的专业 PDF 出题,要求模型跨文本、表格、图表和脚注找证据再回答。
GPQA Diamond 则被直接移出综合指数。这个研究生级科学问答测试已经接近饱和,GPT-6 Astra 最高达到 96.3%,Gemini 3.8 Flash 也达到 95.3%,已经很难区分最强模型。
重新计算后,Claude Fable 5.1 仍排第一,GPT-6 Astra 第二。Astra 得分 55,比 GPT-5.6 Sol 高 4 分。Artificial Analysis 还称,Astra 在同级前沿模型中用了更少的输出 Token。
Disclaimer: The current content is sourced from third-party perspectives or directly translated by AI from third-party perspectives. CoinEx does not guarantee the authenticity, accuracy, and originality of the content, and it does not constitute any investment advice from CoinEx. The prices of cryptocurrencies are highly volatile, please be aware of the potential risks.
- CoinsPrice24H Change