GPT-6.2 Astra scores 20 points above prior best on DeepsecBench security tasks
cramforce · x · 2026-11-18
On Vercel's DeepsecBench, which measures how well models find code vulnerabilities, GPT-6.2 Astra (xhigh) scored 61.22 — roughly 20 points above the previous best — with 56.2% recall, 95% precision, $59.85 cost and 56 minutes runtime; the medium tier scored 53.81 at $30.37. Claude Sonnet 5.5 trails at 37.5. The author argues it's time to rerun AI security reviews with Astra 6.2, as it finds vulns earlier models missed.
Related event: OpenAI's GPT-6.2 Astra Shows Major Gains in Cybersecurity(3 posts)→
More from Models
- Is OpenAI betting on 'good enough'? GPT-6.2 at a fraction of Opus 5.5's price — Bitter-College8786 · 2026-11-18
- Ex-OpenAI safety researchers fear system card safety chatter has gone quiet — j_asminewang · 2026-11-18
- ARC-AGI-3 maintainers explain provider adapter harnesses: vendor-recommended settings per model — mhmazur · 2026-11-18
- Opus 5.5 clones a blogger's writing style in just 30 minutes — IgorCarron · 2026-11-18
- Marin releases all 57 intermediate checkpoints of 535B-A23B MoE trained on 18.2T tokens — dlwh · 2026-11-18
- GPT-6.2 Sol voxel benchmark: mostly a regression vs 6.1, but much faster and cheaper — Angaisb_ · 2026-11-18