GPT-6 Sol/Luna 沿用 Astra 对齐进展
OpenAI 称 GPT-6 Sol 与 Luna 基于 Astra 的对齐成果，在对齐评测上优于 GPT-5.6 同系列模型。
GPT-6 Sol and GPT-6 Luna build on Astra’s alignment advances, showing improvements over the GPT-5.6 Sol and GPT-5.6 Luna models in our alignment evaluations. https://x.com/OpenAI/status/2102460992202420225?s=20
这条是引用着另一条写的：「OpenAI 官方账号发布 GPT-6 Sol 与 Luna 对齐改进」
With improved prompt caching in GPT-6, agents do less redundant processing behind the scenes. More context stays reusable as reasoning effort and tool availability change, making responses faster and cheaper.
On AutomationBench, which measures whether AI models can complete real business workflows across apps, GPT-6 Sol (xhigh) leads Fable 5.1 (max with Opus 5 fallback) with 88% lower reported cost per task.
On DeepSWE v1.1, which tests coding agents on long-horizon engineering tasks, GPT-6 Sol (max) nearly matches Claude Fable 5 (xhigh) at ~80% lower cost per task. GPT-6 Luna (max) is comparable to Fable 5 (medium) at 96% lower cost per task.
Improvements in caching and inference let us deliver GPT-6 Sol and GPT-6 Luna at lower cost. We’re passing those savings on through lower API prices. Sol and Luna deliver more intelligence at half of the previous generation’s API prices, making more complex work practical to automate at scale.
GPT-6 Sol and Luna just landed in Astra’s orbit. Both launch today with API prices 50% lower than GPT-5.6. Build with Sol. Scale with Luna. To production and beyond.