- OpenAI’s GPT-4o now generates high-quality Ghibli-style AI images with improved features.
- Users can enhance results by providing detailed prompts, choosing art styles, and editing images.
- The feature is available for ChatGPT Plus, Pro, Team, and API users, with no watermarks on images.
OpenAI has recently upgraded GPT-4o with improved image generation capabilities, leading to a wave of Ghibli-style AI images flooding the internet. This new feature has excited AI art enthusiasts, allowing them to create high-quality images using simple text prompts.
The AI chatbot works best when users provide highly specific prompts. Instead of vague descriptions, adding details about the background, colors, and rendering style can significantly improve the results. For example, asking for a “hyper-realistic sunset with ocean reflections and birds in the horizon” gives better results than just “a sunset.”
Another key feature is the ability to generate images in different artistic styles. Apart from Ghibli, ChatGPT can create images in voxel, lo-fi, anime, oil painting, and more. Users can also modify existing images by instructing ChatGPT to change colors, add objects, or edit specific details.
The aspect ratio of images can also be adjusted based on user needs. While ChatGPT typically generates square images, users can request different ratios like 16:9 for desktops, 9:16 for mobile wallpapers, or 1:1 for profile pictures.
Though AI chatbots struggle with adding detailed text to images, GPT-4o can place simple phrases like “Happy Birthday” or “Get Well Soon.” This makes it useful for creating quick greeting cards or social media posts.
The improved image generation feature is available for ChatGPT Plus, Pro, Team, and API users. Unlike DALL-E, the images from GPT-4o do not have a watermark. The ChatGPT Plus plan, which includes this feature, is available for Rs 1,999 per month.
ALSO READ: “OpenAI Announces New ‘Deep Research’ Tool For ChatGPT Amid DeepSeek Buzz”
Comments