Perplexity trusts GPT-6 Astra with end-to-end systems
OpenAI announced on September 14, 2026, that Perplexity is trusting GPT-6 Astra for end-to-end system testing. This model can generate realistic responses that mimic those from services like language model APIs or connectors. By simulating these services, GPT-6 Astra allows Perplexity to evaluate how its applications respond and to test entire workflows from beginning to end, ensuring comprehensive system validation.
This marks the first time Perplexity has publicly committed to using GPT-6 Astra for end-to-end system testing, unlike previous, more limited integrations.
时间与来源
时间显示为 UTC
显示时区：UTC
本地时区尚不可用，暂时显示 UTC。
发布当时偏移：UTC+02026年9月13日 15:00 UTC
收录当时偏移：UTC+02026年9月12日 02:01 UTC
- 发布
- 2026年9月13日 15:00
- 收录
- 2026年9月12日 02:01
- 来源类型
- 官方发布
- 档位
- 当事方
- 信源状态
- 正常
档位是按信源手工设定的编辑判断，不是逐条打分。
讨论趋势
百分比基于采集到的讨论信号，不代表新增评论数或独立参与人数。曲线仅用于同一话题在不同时段的比较。
OpenAI September 14, 2026
Perplexity uses Astra to write communications, change software, and monitor production systems, and checks in much less frequently than with earlier models.
Company size: Startup
Region: North America
Industry: Technology
Products: API
As an AI-powered answer engine, Perplexity is deeply focused on search and accuracy. Its ability to process large amounts of information is critically important. Johnny Ho, Cofounder and Chief Strategy Officer, observes that every time the model gets better at writing code, Perplexity’s search engine improves too. It becomes able to write better programs that search the web and internal information and summarize it very concisely.
But the real challenge, according to Johnny, is taking those informational aspects and applying them to real-world systems. Something made easier with GPT‑6 Astra.
“We can have the model craft communications, edit real-world systems, and monitor our production software in a way that previous generations were not able to.”
—Johnny Ho, Cofounder and Chief Strategy Officer, Perplexity
Letting the model do the testing
For Johnny, one of the most useful applications of AI is testing code. With limited time to test manually, he asks GPT‑6 Astra to build a small testing program around an application.
The model generates realistic responses like those another service would send, for example, a language model API or a connector. By standing in for those services, the model can check how the application responds and test the workflow from start to finish.
“We’re actually able to trust it with full end-to-end systems and check in on it much less frequently than previous generations of models.”
Keep reading