deepseek-v4.1IMAGE
渗透测试基准：V4.1 Flash 1017 找到 316 个漏洞，成本仅 $32.93
在 8 个真实 Web 应用上，Flash 1017 的 F1 为 75.08、花费 $32.93，高于 GLM-5.3 的 73.53/$305、GPT-5.6 Sol 的 69.80/$1983 和 Opus 5 的 70.12/$3032。
JJOKAH
@JohnJohnsonOkah3 天前收录原推发布于 2026.10.26
JJOKAH @JohnJohnsonOkah原推发布于 2026.10.26
Open-weight model (DeepSeek) outperformed every frontier model on a pentesting benchmark. DeepSeek V4.1-Flash-1017 - found 316 known vulnerabilities across 8 live web apps. F1 score / total cost: - DeepSeek V4.1-Flash: 75.08 / $32.93 - GLM-5.3: 73.53 / $305.14 - GPT-5.6 Sol: 69.80 / $1,983.18 - Claude Opus 5: 70.12 / $3,032.39 In Penetration Testing, open models are now taking the lead and no longer the cheaper alternative.
x.com/johnjohnsonokah/status/2114560053910933615采集于 2026.10.26 12:03