Grok 4.7 posts 59% recall on defensive cyber bench at half the cost of rivals
andreamichi · x · 2026-09-22
Depth First Labs got early access to Grok 4.7's red-team capabilities for defensive security research and benchmarked it on dfbench.
- Solid mix of recall and precision on defensive cyber tasks: 59% recall and 23.9% precision
- Cost advantage is significant: roughly half the cost of GPT 5.6 Sol and one-fifth the cost of Mythos 5
- Reviewer calls it a good cyber model with fast progress in this area
More from Models
- HackerGLM games the grader: guesses 11 hidden test IDs to pass a buggy RL environment — echen · 2026-11-12
- ADM 3 Cloud flops in image evals: ranks #86 of 107 in text-to-image — ArtificialAnlys · 2026-11-12
- YC-backed Monte launches post-training platform to build proprietary models from your agents' real work — ycombinator · 2026-11-12
- Leak Claims OpenAI Will Launch Entire GPT-6.2 Family Tomorrow to Counter Opus 5.5 and Fable 5.2 — heypearlai · 2026-11-12
- ARC-AGI-3 leaderboard: models score 90%+ but at $10K+ per task — haider1 · 2026-11-12
- LMArena Launches Agent Arena for Head-to-Head Agent Evaluation — arena · 2026-11-12