OpenAI 发布 GPT-6.1号机车。 以半价价获得更快，表现更好的AI agents
OpenAI Group PBC今天的发布 GPT-6.1 Terra，是其前沿人工智能系统阵列中的中层模型的升级。
新的模式是GTP-5.6 Terra的继任者，在更快和更精确地处理计算机使用任务时，成本大约减少50%。
Terra被定位为OpenAI的日常模式，设计用于日常工作，如读取电脑屏幕上的内容、点击应用程序、浏览网站和起草信件。 50%的价格削减适用于该模型的标准应用编程接口率，这是开发者为了将GPT-6.1 Terra整合到自己的应用中而支付的费用。
OpenAI认为，它是一种折衷模式，对于填写表格、解释技术图纸、编辑视频、制作图像和自动化重复任务等更简单的任务来说，它的速度更快、成本更低。
OpenAI在介绍该模式的博客文章中写道，“从GPT-5.6 Terra到workday的通讯和计算机任务，从起草提案和解释技术图纸到填写表格和编辑录像，取得了显著进展。”
GPT-6.1 Terra：以较低的价格提供快速且有能力的计算机使用。
GPT-6.1 Terra了解屏幕上的内容，图片。 twitter.com/8JTN1yonmV
— OpenAI (@OpenAI) 2026年10月26日 (中文(简体) ).
OpenAI 称该型号现在可以提供给ChatGPT Plus,Pro,Enterprise和Edu所有ChatGPT Work和Codex的用户。 ChatGPT 电话工作是agent模式在ChatGPT内部的功能，旨在代表第三方应用程序和浏览器中的用户完成工作，而Codex是OpenAI的软件编码工具。
对GPT-6.1 Terra的一个关切是，其较先进的能力在安全方面似乎略有损失。 OpenAI 称在其博客文章中表示，在近30%的模拟运行中，Terra违反限制
此外，OpenAI透露，GPT-6.1 Terra在遇到来自其他AI agents的消息后，在13.75%的使用案件中采取了未经授权的行动。 该公司的系统卡还有显示，研究者发现了该模型理解其正在测试的证据。
尽管如此，OpenAI 称 GPT-6.1 Terror在一些衡量标准上表现得更好例如，在利用旨在欺骗模型犯错的提示建立的内部测试中，低推理能力的事实错误率从GPT-5.6 Terra的53.6%下降到16.7%。 在作出最大努力时，其事实错误率降至4.3%。
在对通过计算机指令线终端完成的任务进行编码的终端-Bench 4.0测试中，它用最大推理能力解决了50.3%的所有任务，而用GPT-5.6 Terra解决了21.5%的任务。 OpenAI 称该模型平均完成这些任务的时间为3次，快于GPT-6.1 Sol和GPT-6 Astra等更强大的变体。
Terra的发行跟随GPT-6.1 Sol上个月的首播。 线条还包括一种轻量级模型变体，称为GPT-6.1 Luna.
图片：OpenAI
SiliconANGLE联合创始人John Furrier的留言：
支持我们的使命，通过与CUBE社区接触，保持内容的开放和自由。 加入CUBE的校友信托网络，其中技术领导者相互连接，分享情报，创造机会。
- 15M+观看CUBE视频，为AI、云层、网络安全等各种对话提供动力
- 11.4k+CUBE校友——通过独特的信任网络与11400多名技术和企业领袖建立联系，塑造未来
你是AWS客户吗？在财务上支持SiliconANGLE，从我们的市场门户页面和链接购买您的AWS服务：https://siliconangle.com/aws-marketplace/.
关于 SiliconANGLE 媒体
SiliconANGLE Media由技术视觉家约翰·弗瑞尔(John Furrier)和戴夫·韦兰特(Dave Vellante)创立，构建了产业领先数字媒体品牌的动态生态系统，达到1500+百万精英技术专业人士。 利用CUBEAi.com神经网络，帮助科技公司以数据为导向做出决策
本文由 AI图鉴自动抓取并翻译，版权归原作者所有；如需查看原文请点下方链接。
OpenAI launches GPT-6.1 Terra for faster, better-performing AI agents at half the price
OpenAI Group PBC today released GPT-6.1 Terra, an upgrade to the middle tier model in its lineup of frontier artificial intelligence systems.
The new model, the successor to GTP-5.6 Terra, costs around 50% less while handling computer use tasks faster and with greater accuracy.
Terra is positioned as the everyday model in OpenAI’s lineup, designed for routine work such as reading what’s on a computer screen, clicking through applications, navigating websites and drafting messages. The 50% price cut applies to the model’s standard application programming interface rates, which is what developers pay to integrate GPT-6.1 Terra into their own applications.
According to OpenAI, it’s meant to be a kind of compromise model that’s faster and cheaper for simpler tasks such as filling out forms, interpreting technical drawings, editing videos, creating images and automating repetitive tasks – work that doesn’t require the heft of its more advanced model, GPT-6.1 Sol.
“We’re introducing GPT-6.1 Terra, a fast and efficient model that combines detailed visual reasoning and precise computer use alongside clearer, more natural writing,” OpenAI wrote in a blog post introducing the model. “A significant step up from GPT-5.6 Terra, it excels at the communication and computer-based tasks that fill the workday, from drafting proposals and interpreting technical drawings to filling out forms and editing videos.”
GPT-6.1 Terra: delivering fast and capable computer use at a lower price.
GPT-6.1 Terra understands what’s on screen, takes action across apps, writes naturally, and is a more capable model for your whole workday with the efficiency to take on more of it. pic.twitter.com/8JTN1yonmV
— OpenAI (@OpenAI) October 26, 2026
OpenAI said the model is available now for all ChatGPT Plus, Pro, Enterprise and Edu users in ChatGPT Work and Codex. ChatGPT Work is the agent mode feature within ChatGPT, designed for completing work on behalf of users in third-party applications and browsers, while Codex is OpenAI’s software coding tool.
One concern with GPT-6.1 Terra is that its more advanced capabilities appear to come at a slight cost in terms of safety. In its blog post, OpenAI said that in nearly 30% of simulated runs, Terra went against restrictions to complete tasks using methods it had no authorization to use.
In addition, OpenAI revealed that GPT-6.1 Terra took unauthorized actions in 13.75% of use cases after coming across messages that came from other AI agents. The company’s system card also shows that researchers found evidence that the model understood it was being tested.
Nonetheless, OpenAI said GPT-6.1 Terra did better in some metrics. For instance, on an internal test built using prompts designed to trick models into making mistakes, its rate of factual errors at low reasoning effort fell to just 16.7%, down from 53.6% with GPT-5.6 Terra. When making maximum effort, its rate of factual errors dropped to just 4.3%.
On the Terminal-Bench 4.0 test, which measures coding agents on tasks done through a computer’s command-line terminal, it solved 50.3% of all tasks when using maximum reasoning effort, up from 21.5% with GPT-5.6 Terra. OpenAI said the model completed these tasks around three-times faster on average than more powerful variants such as GPT-6.1 Sol and GPT-6 Astra.
The release of Terra follows the debut of GPT-6.1 Sol last month. The lineup also includes a lightweight model variant called GPT-6.1 Luna.
Image: OpenAI
A message from John Furrier, co-founder of SiliconANGLE:
Support our mission to keep content open and free by engaging with theCUBE community. Join theCUBE’s Alumni Trust Network, where technology leaders connect, share intelligence and create opportunities.
- 15M+ viewers of theCUBE videos, powering conversations across AI, cloud, cybersecurity and more
- 11.4k+ theCUBE alumni — Connect with more than 11,400 tech and business leaders shaping the future through a unique trusted-based network
Are you an AWS customer? Support SiliconANGLE financially by buying your AWS services from our Marketplace portal page and links: https://siliconangle.com/aws-marketplace/
About SiliconANGLE Media
Founded by tech visionaries John Furrier and Dave Vellante, SiliconANGLE Media has built a dynamic ecosystem of industry-leading digital media brands that reach 15+ million elite tech professionals. Our new proprietary theCUBE AI Video Cloud is breaking ground in audience interaction, leveraging theCUBEai.com neural network to help technology companies make data-driven decisions and stay at the forefront of industry conversations.
Auto-fetched and translated by AI Atlas. All rights belong to the original author.