XBOW evals: GPT-6 vs Claude Mythos 5.1 exploit success per dollar on 65 vulnerable apps
mboehme_ · x · 2026-11-10
oegerikus and XBOW tested Claude Mythos 5.1 and OpenAI's GPT-6 lineup on 65 vulnerable open-source web apps (RCE, XSS, SSRF, SQLi and more) to measure exploit success per dollar.
The XBOW Model Runtime Evals (2026-11-09) spend-success curves show gpt-6.1-sol is most cost-efficient at low spend (75%+ success around $1), while claude-mythos-5.1 and gpt-6-astra need $10 to match; gpt-6-luna leads early but plateaus near 66%; all models eventually approach 95%.
More from Models
- DeepMind's Samuel Albanie: fable 5.2 is SOTA for clarity on technical topics — SamuelAlbanie · 2026-11-10
- Hands-on: Tencent's Hy4 matches or beats rivals across five agentic coding tests — CodeByPoonam · 2026-11-10
- Unverified rumor: Gemini 4 Carbon rumored to launch today as UGI board shows Argon tiers — Eissa_Cozorav · 2026-11-10
- Max/Ultra Subscriber Burns Through 5-Hour Quota in 1 Hour, Slams Claude Inference Capacity — theteknosaur · 2026-11-10
- Ant's Ling-3.2-Flash Adds JSON Schema Structured Output Support — Wonderful-Switch3107 · 2026-11-10
- Fable 5.2 reported nearly unusable in Cursor over context truncation — IndraVahan · 2026-11-10