AITOPAI热报
ModelsMulti-sourceSelected

新模型在 CVE 漏洞复现基准上追平 DeepSeek V4 Pro 0813

Teortaxes@teortaxesTex
Post· 10 sources
View the post
Why we picked it

一个模型在漏洞复现测试里找回 87.5% 的 CVE，跟 DeepSeek V4 Pro 0813 打平，只差 GPT-6 Astra 三个百分点，安全研究方向可以看看。

在 32 个基准 CVE 的复现测试中，该模型以 pass@3 找回 28 个，比例 87.5%，与 DeepSeek V4 Pro 0813 并列第二。榜首是 GPT-6 Astra，成绩为 90.6%。发帖人指出 DeepSeek V4 Pro 0813 的成绩不可能是从 Astra 蒸馏得来，认为其优势来自数据规模。

Original · Teortaxes

> - At pass@3, it rediscovered 87.5% of the benchmark CVEs (28/32). That ties DeepSeek V4 Pro 0813 for 2nd place, behind only GPT-6 Astra (90.6%)

Honestly, crazy that 0813 already did that and no, they couldn't have distilled Astra for 0813, this is Whale Data Supremacy

#DeepSeek#GPT-6 Astra#漏洞复现#推理模型

Keep reading

Recent stories with the same tag or section
ProductsUS vs ChinaSelected12:24
DeepSeek Search ranks 17th on Artificial Analysis Search Index

The headline price is low, but the benchmark shows the real cost lives in tokens: ~7k-token search results make DeepSeek's model cost the 2nd highest on the board at ~US$0.023 per task. Developers building search-augmented agents should weigh that against the weak BrowseComp score of 63%, which trails Perplexity variants and Octen at 85% to 87% on multi-hop tasks. Compared with alternatives like Exa and Firecrawl, DeepSeek wins on price, not on factual accuracy.

#DeepSeek#Search API#Artificial Analysis#agents
官方账号Artificial Analysis@ArtificialAnlysSource