Xbow on offensive security model evals: the best model depends on your budget
moyix · x · 2026-11-07
Xbow shared a thread on its latest frontier model evaluations for offensive security, arguing that the best model depends on how much you're willing to spend. The post is only the thread opener; detailed scores and cost figures are in the original thread.
More from Models
- RareDiseaseBench: AI agents max out at 43.8% on 35 rare-disease diagnostic workups — kenbwork · 2026-11-07
- VIALS benchmark: frontier models flunk 161 life-sciences image interpretation tasks — niloofar_mire · 2026-11-07
- MIT's Drift Calibrator visualizes sycophancy and empathy drift live during AI chats — patpat_mit · 2026-11-07
- Estimated Anthropic internal model scores show only ~4.5-point AECI gain across three generations — ChrisGPT · 2026-11-07
- Kimi K3.1 Pro lands on Tinker with 256K context, its largest open-weights model — _kevinlu · 2026-11-07
- DeepMind researcher: next-gen model gains come from bigger pretrain + data/algorithm gains, not compute speedup — tobyordoxford · 2026-11-07