OpenAI Launches GPT-6 Astra: 88% on Reverse-Engineering Benchmark, 1M Context Accuracy Jumps to 96.3%
rickasaurus · x · 2026-09-04
OpenAI officially launched GPT-6 Astra, pitched as "anything you can do on a computer, Astra can do for you. Fast." Key details:
- Reverse engineering: on SRE-Bench (reverse-engineering binaries without source code), Astra solved 88.0% of tasks in one attempt and 99.2% within four, vs 55.9% and 68.7% for GPT-5.6 Sol.
- Long context is a huge upgrade: OpenAI MRCR v2 8-needle improved from 91.5% to 100.0% at 256K-512K, and from 73.8% to 96.3% at 512K-1M — making 1M-token context windows actually usable instead of silently dropping 9-30% of context.
Commenters argue these two gains effectively "solve" reverse engineering and turn million-token context from a spec sheet number into a practical capability.
Related event: OpenAI Launches GPT-6 Astra, Declaring 'Welcome to the AGI Era'(159 posts)→
More from coding & agent
- Largest open experiment finds dev tools are always mentioned but never chosen by coding agents — ycombinator · 2026-09-04
- Enterprise AI Agent Cohort Kicks Off With Engineers From PlayStation, Salesforce, ServiceNow — hugobowne · 2026-09-04
- Mathematician tests GPT-6 Astra: live Lean proof verification while writing arguments — teortaxesTex · 2026-09-04
- Makepad flow: a Rust single-executable ComfyUI alternative built in one day — anselm · 2026-09-04
- antirez: judge new models by whether they fix real blocking bugs, not three.js demos — antirez · 2026-09-04
- Getting an MCP App running inside Gemini Enterprise to surface Cloud Run services — rseroter · 2026-09-04