跳到正文
原文
eric zakariasson· @ericzakariasson · X·· 9 天前AI 评分75
AI 导读

Eric Zakariasson 分享 Grok 4.7 模型卡,列出多项基准相对 4.6 的提升:Terminal-Bench 20.3% → 38.0%,SWE-Marathon 31.9% → 46.0%,HealthBench Pro 48.5% → 56.7%,Legal Agent 15.8% → 19.6%,EEBench 60.0% → 66.0%。作者称价格与 4.6 相同,并附上模型卡 PDF 链接 https://media.x.ai/v1/website/4p7card-5eccc980.pdf。

正文

and here’s the grok 4.7 model card. a few jumps vs 4.6 that stood out:

- Terminal-Bench: 20.3% → 38.0%
- SWE-Marathon: 31.9% → 46.0%
- HealthBench Pro: 48.5% → 56.7%
- Legal Agent: 15.8% → 19.6%
- EEBench: 60.0% → 66.0%

for the same price as 4.6!

https://media.x.ai/v1/website/4p7card-5eccc980.pdf

来源:eric zakariasson · x.com