Payal Manghnani's Post on Medial | Join the hottest discussion on ...
OpenAI might introduce a new image generation tool in ChatGPT soon. This tool includes a "thinking" phase before generating images and can generate images in multiple stages. It is speculated that there may be a standard image generation process and an 'XL image' mode for more detailed outputs. The process itself could take around 30 seconds or more to complete.
Reasoning in Images
Researchers at Google DeepMind introduced Semantica, an image-conditioned diffusion model capable of generating images based on the semantics of a conditioning image. The paper explores adapting image generative models to different datasets.
ChatGPT's Image Generation Feature
ChatGPT now offers direct image generation! Users can now create images directly within ChatGPT, powered by the impressive DALL-E 3. Simply describe what you want, and even make adjustments through text prompts. However, it's important to note that only two image genera...
DeepSeek's Janus Pro 7B Release
DeepSeek has released open-source Janus Pro 7B for image understanding and generation! This model boasts several impressive statistics such as SOTA 0.8 on GenEval and 84.19 on DPG-Bench, beating DallE3 and SD3-Medium. It has 72 million synthetic images in pretraining with good text rendering capabilities. However, it is noted that images generated are small at 384x384 resolution.
AprameyaAI's Funding for GenAI
AprameyaAI has received funding for integration into Fintech products. This funding will optimize the prompt engineering assistant designed for generative AI (GenAI) applications. The aim is to simplify the process of generative AI applications.
Free AI Tools and Image Generation
A list of favorite free AI tools is provided including Gemini 1.5 Pro, which allows users to upload audio, video files, or PDFs and ask questions related to them. Additionally, Claude and Microsoft Copilot are mentioned for their capabilities.
Midjourney Updates
Midjourney has released version 6.1 with new features including more coherent images, improved image quality, and better precision and details in image generation.
