OpenAI宣布ChatGPT免费用户每天可用DALL-E 3模型生成至多两张图像，该模型的关键改进是ChatGPT可生成制作图像的提示，使其更易于生成图像，且已开始逐步推出，作者在写作过程中已进行尝试。

🎨OpenAI推出新举措，ChatGPT免费用户可借助DALL-E 3模型每日生成最多两张图像。此模型的一项重要改进是，ChatGPT能够给出生成图像的提示，这使得图像制作变得更为简便。例如，可让ChatGPT为幻灯片制作图片、为朋友个性化设计卡片或展示某物的外观等。

🌟在演示中，DALL-E团队的首席研究员兼负责人Aditya Ramesh通过ChatGPT生成一个山区拉面馆的标志提示，ChatGPT随后写出一个更长的提示，DALL-E则提供了四个选项，展示了该模型的实际应用效果。

🚀OpenAI表示用DALL-E 3创建图像的功能正在逐步推出，作者在撰写本文时，通过ChatGPT Mac应用程序成功生成两张图像后收到了达到当日图像创建限制的通知，证明该功能已在部分场景中实际应用。

OpenAI is going to let users of ChatGPT’s free tier make up to two images per day with its DALL-E 3 model, the company announced on Thursday. When the company launched DALL-E 3 in September, it was available first to users who paid for ChatGPT Plus.

One of DALL-E 3’s key improvements is that ChatGPT can come up with a prompt to make an image, which should make it easier to make images. For example, as my former colleague Emilia David wrote last year:

In a demo to The Verge, Aditya Ramesh, lead researcher and head of the DALL-E team, prompted ChatGPT to help him come up with a logo for a ramen restaurant in the mountains. ChatGPT then wrote a longer prompt, and DALL-E came up with four options.

OpenAI says the ability to create images with DALL-E 3 is “rolling out,” but you might already have access: while writing this article, I was able to make two images with the ChatGPT Mac app before getting a notice that I had reached my image creation limit for the day.

It’s been a busy day for OpenAI news. The company released a safety assessment of its GPT-4o model, added a new person to its board of directors, and CEO Sam Altman was sent a letter from Democrats in Congress pushing for answers about OpenAI’s safety record.