Tag
OpenAI 三个名为 aurora-red-alpha、aurora-green-alpha 和 aurora-blue-alpha 的新图像模型引用疑似在 API 与 Arena 盲测中意外泄露，外界猜测它们可能就是 GPT-Image-3。
The author shares a complete workflow for building 3D models from scratch with AI: first use GPT Image to generate front, back, left, and right views, then pass them to TripoAI to generate the model, split the parts, apply textures, retopologize, rig, and export — finally, use GPT to generate a web preview page.
A user shares how to use GPT Image 2.5 to generate rose champagne pink CCD-style images with detailed prompts for East Asian fashion scenes.
A tweet from @calicastle praises the perfection of GPT image 2.5 sunburst for creating app store screenshots.
A Twitter thread comparing the differences between GPT Image 2.0 and 2.5 using challenging prompts from ImagineArt.
User @corbin_braun plans to upgrade YouTube thumbnails using GPT Image 2.5 and Thumio W, promising a fundamental enhancement.
A user highlights that GPT image 2.5 shows excellent character consistency in image generation.
GPT Image 2.5 的新素描功能令人惊艳，能将简单笔触转化为极具创意和美感的图像，展示了AI的强大想象力。
The tweet opines that GPT Image 2 is endgame and questions the necessity of GPT Image 3.
Corbin Braun suggests that OpenAI's updated GPT Image 2 is a game-changer for thumbnails, having created one with a single prompt in Thumio.
A tweet claims OpenAI silently shipped a massive upgrade to GPT Image 2, producing impressive results with a single prompt, while promoting an AI thumbnail tool.
A potential new GPT-Image model, reportedly named "Mona-lisa-1", has appeared on the LMArena, with OpenAI SynthID watermarks detected in its outputs.
A user compares four image generation models—Qwen Image 3.0, GPT Image 2, Nano Banana 2, and Seedream 5.0 Pro—with four prompts in a Twitter thread.
Share a set of prompts for GPT-image2 and Gemini that automatically generate English word learning cards in Cambridge/Oxford style, including IPA, definitions, example sentences, illustrations, etc.
Sam Altman comments that a user's demo of using an AI agent to analyze wardrobe photos and generate new outfit renderings with GPT-image would have been a whole startup not long ago.
The user released an open-source AI illustration skill called guizang-material-illustration, based on GPT-Image 2.0 and Codex Agent, which can generate 3D material explanation diagrams with Chinese labels, suitable for articles, weekly reports, PPTs, etc.
Google Antigravity releases a 19-minute tutorial on building an award-winning animated website using Antigravity and GPT Image 5.5.
GPT Image 2.0 demonstrates remarkably strong comprehension capabilities, understanding verbose prompts from a self-built director agent and generating decent results, highlighting the evolving role of filmmakers in the AI era.
A new demo called Reelquest uses Seedance 2.0 and GPT Image 2 to create a fully interactive, playable game where every frame is AI-generated based on user actions, requiring no coding or game engine.
The user shares an impressive example of a 3D biological structure page generated using GPT Image 2 and Gemini 3.1 Pro, noting its potential for AI education and expressing intent to replicate it.