R125 · Infrastructure for agents at scale › Frontier model releases

GPT-6.2 Astra and Sol: cheaper frontier tasks, and the first High rating for AI self-improvement

Read the original

model-releasebenchmarkspricingcachingai-safety

Source: OpenAI, Introducing GPT-6.2 Astra and Sol (2026-11-17), announced on LinkedIn the same day, with a linked system card. All scores below are vendor-reported launch numbers; treat them as claims until independent runs appear.

What shipped

The headline is cost per task more than raw score: OpenAI attributes most of the saving to fewer output tokens, plus the price cut.

Claimed results

Safety and capability thresholds

Engineering implications

Review checklist before switching

Connections

See R125 on the knowledge map

Part of the Research notes on nowucca.com. Figures and claims are the source's own unless the summary says otherwise.