AI 导读
每个问题都属于三种类型之一：从列表中选一个选项、按有序等级对上下文打分，或返回某陈述为真的概率。 你可以在单次调用中包含数十个问题，每个问题独立评估，以稳定响应时间。 Python 和 JavaScript SDK、浏览器 playground，以及面向 Claude Code 和 Codex 的 agent skill 均已上线。早期 API 访问通过开发者 waitlist 进行。 来试试看 👀 https://t.co/SKz6xiG2f4
正文
Every question comes in one of three types: pick one option from a list, score the context against ordered levels, or return the probability that a statement is true.
You can include dozens of questions in a single call, each evaluated in isolation, to stabilize response time.
Python and JavaScript SDKs, a browser playground, and an agent skill for Claude Code and Codex are live. Early API access runs through a developer waitlist.
Test it out 👀
https://t.co/SKz6xiG2f4
来源：@testingcatalog · x.com