OpenAI 推出了其最新、最强大的 AI 模型 GPT-6 Astra，该模型具有新颖的“计算机使用”能力，可以像人类一样在计算机上导航。该新模型在数学能力、软件工程和网络安全等领域取得了显著进步，在各种基准测试中表现优于之前的 OpenAI 模型和 Anthropic 的 Claude Fable 5.1 等竞争对手。尽管功能先进，Astra 的发布是分阶段进行的，首先面向网络安全计划中的特定企业客户，随后计划向 API 和云用户更广泛地提供，而高级网络安全任务将受到限制。
AI
影响
在计算机使用和网络安全基准测试上设定了新的 SOTA（State-of-the-Art），可能加速 AI 代理能力并引发安全担忧。
new SOTA computer use and coding, 2.5x pricier per token, but WAY cheaper per task, less monitorable. overall, a very successful launch of OpenAI’s new frontier model class.
OpenAI president Greg Brockman calls Astra "the start of the AGI era" even as outside experts worry about whether the company has effective safety and security controls in place.
<h2> OpenAI Dropped GPT-6 Astra. Here's What the Benchmarks Actually Say (and Don't Say) </h2> <p>OpenAI released GPT-6 Astra on September 3, 2026 — the first full-number release since GPT-5. It hits 99.9% on ARC-AGI-3 with the Provider Adapter harness, improves token efficiency …
<table> <tr><td> <a href="https://www.reddit.com/r/singularity/comments/1w5f55h/gpt6astra_has_been_staged_on_the_openai_api/"> <img alt=""GPT-6-ASTRA" has been staged on the OpenAI API" src="https://preview.redd.it/qa0i0dfyu4nh1.png?width=140&height=140&crop=1:1…
