OpenAI releases GPT-6.2 Astra and Sol amid mixed early reviews
OpenAI has launched two new models, GPT-6.2 Astra and GPT-6.2 Sol, positioning Astra as its top model for complex professional and scientific work while Sol targets everyday tasks at lower cost. OpenAI says Astra hits new benchmark highs at under a third of its predecessor's per-task cost, but early testers reported issues including unproductive loops, premature task completion, and confusing outputs.
GoKawiil's interpretation of the reporting above, not reported fact.
The gap between benchmark performance and real-world usability suggests efficiency gains may come with tradeoffs in reliability, which matters as OpenAI pushes models toward handling entire projects via tools like ChatGPT Work rather than just answering questions. OpenAI's suggestion that some reviewers missed late improvements indicates the company is still refining the models' behavior even after release.
- OpenAI launched GPT-6.2 Astra for complex professional and scientific tasks, and GPT-6.2 Sol for everyday work.
- Astra reportedly achieves new benchmark highs at less than a third of its predecessor's cost per task.
- Early testers flagged problems like unproductive loops and premature task completion despite strong benchmarks.
Source: androidauthority.com, 2026-11-18
Published there as: “OpenAI launches GPT-6.2 Astra and Sol, but not everyone is sold on the cheaper upgrade”
Read the original report → The summary and analysis above are GoKawiil's own, written from reporting by the source above. Facts and quotes belong to the original publisher.