So, the ChatGPT numbers for remaining Agent Mode and Deep Research can't be right. I haven't used either this billing cycle yet, and I've got uses missing. 🤔
I’ll give Grok Imagine some credit for instruction adherence, but the frame rate is awful, movements of the characters often look unnatural or the model just resorts to shaking the camera, and the audio almost never matches the sequencing of the motion. Sometimes the audio is
Actually amazing if this were true in-painting. Too many details of the edited images are changed along with the instructions followed. Get that right along with the Grok Imagine speed would actually be quite useful.
I’ve been waiting for Paradigm to be generally available for a while. Very cool stuff enabled here, and it feels like a great competitor to Exa Websets.
🚀 Excited to introduce Qwen-Image-Edit!
Built on 20B Qwen-Image, it brings precise bilingual text editing (Chinese & English) while preserving style, and supports both semantic and appearance-level editing.
✨ Key Features
✅ Accurate text editing with bilingual support
✅