Event ·Multi-source

网络攻防基准：开源模型 DeepSeek V4.1 Flash 追近 GPT-6

Vals AI 基准测试显示，OpenAI 的 GPT-6 Sol Daybreak 在模拟企业网络的渗透测试中达成 80.8% 的攻击目标。开源权重模型 DeepSeek V4.1 Flash 达到 72.8%，成本仅为前者的十分之一。分析人士指出，开源模型无法被限制访问，防御方的应对窗口正在快速缩小。

Current read

Vals AI 测了 GPT-6 和 DeepSeek V4.1 的网络攻击能力，开源模型花十分之一的钱就追到只差 8 个百分点，做安全的得看看。

11 sources57° AI heatLast updated 2026/11/7 15:35:50

Evidence

11 sources

Primary sourceDecoder

2026/11/7 15:35:50View source

Related source 1TestingCatalog

2026/11/5 17:12:21View source

Related source 2Greg Brockman

2026/11/5 17:31:11View source

Related source 3rohanpaul_ai

2026/11/5 22:27:59View source

Related source 4theverge

2026/11/5 17:00:00View source

Related source 5OpenAI

2026/11/5 17:09:13View source

Related source 6Andrew Curran

2026/11/5 17:25:52View source

Related source 7Tech in Asia

2026/11/6 00:28:44View source

Related source 8Blognone

2026/11/6 09:46:37View source

Related source 9lmarena.ai

2026/11/6 22:43:27View source

Related source 10kimmonismus

2026/11/5 17:21:00View source

Conflict check

Dedup data only shows that these sources discuss the same event, not that they agree. There is no structured supporting or refuting evidence yet, so no conflict is inferred automatically.

Open the story