6 家媒体·9 篇报道·历时 35m
ChatGPT 图像 2.0 发布
OpenAI 推出了 ChatGPT 图像 2.0,提供即时(Instant)和思考(Thinking)两种模式,可生成最高 2K 分辨率、不同纵横比的图像。该模型在渲染日语、韩语、印地语和孟加拉语等非拉丁文字方面表现更强,能够生成文本、信息图、幻灯片、地图和漫画。在思考模式下,它还能浏览网页,从单一提示生成多幅图像。
本文综合以下全部报道写成，而非依据任何单一来源。
本页由机器翻译生成，原始报道为其原文语言。
各家媒体如何报道
- Hacker News·ChatGPT Images 2.0
- Wired·OpenAI Beefs Up ChatGPT's Image Generation Model
- VentureBeat·OpenAI's ChatGPT Images 2.0 is here and it does multilingual text, full infographics, slides, maps, even manga — seemingly flawlessly
- Engadget·ChatGPT Images 2.0 is better at rendering non-Latin text
- TechCrunch·ChatGPT’s new Images 2.0 model is surprisingly good at generating text
- TechMeme·OpenAI releases ChatGPT Images 2.0 with new "thinking capabilities", allowing it to search the web to help it create multiple images from a single prompt (Emma Roth/The Verge)
- TechMeme·OpenAI says that ChatGPT Images 2.0 has a stronger understanding of non-Latin text rendering in languages like Japanese, Korean, Hindi, and Bengali (Amanda Silberling/TechCrunch)
- TechMeme·OpenAI says ChatGPT Images 2.0 comes in Instant and Thinking variants and can generate images of up to 2K resolution and in multiple aspect ratios (Zac Hall/9to5Mac)
- TechMeme·ChatGPT Images 2.0 is available globally to ChatGPT and Codex users, with a more powerful version for paying subscribers; its knowledge cutoff is December 2025 (Reece Rogers/Wired)