OpenAI's Astra scores 100% on ExploitBench, discovers zero-days in internal test
zephyr_z9 · x · 2026-09-02
OpenAI's new model Astra demonstrates a significant leap in cybersecurity capabilities. It scored 100% on the ExploitBench benchmark, surpassing Sol-5.6 (Max) at 73.5% and Mythos 5 at 78%. To prevent benchmark contamination, OpenAI tested it on a new internal benchmark with recent vulnerabilities. With roughly comparable output tokens (77k), Astra scored 39% while GPT-5.6 Sol scored only 1%. During evaluation, Astra also discovered and utilized two zero-day vulnerabilities as part of an exploit chain.
More from Models
- Shopify's 0.8B fine-tune beats GPT 5.6, saves millions — CShorten30 · 2026-09-02
- Rumor: Gemini 3.8 Flash Arrives Tomorrow, Skipping 3.5 Pro — Hesamation · 2026-09-02
- Claude Fable 5.1 Replicates and Extends Research; Mythos Writes Custom Kernels — BenBlaiszik · 2026-09-02
- Developer finds flaws in CritPt benchmark scores — scaling01 · 2026-09-02
- MiniMax-M3 hits 2,274 tok/s in Cacheon arena, focused on enterprise end-to-end testing — const_reborn · 2026-09-02
- Claude flags garden keep-out zones, suggests reorienting camera for safety — _Stocko_ · 2026-09-02