Weeks after shutting down its Sora project, OpenAI is reportedly developing a new image generation model called gpt-image-2, which aims to produce more realistic visuals compared to its competitors. This initiative follows the recent popularity of ChatGPT’s image generation tools, which have gained significant traction for their ability to create stylized graphics, including anime-inspired artwork.
The new gpt-image-2 model is designed to address criticisms of previous AI-generated images, which often exhibited overly smooth textures, unnatural lighting, and unrealistic facial features. According to sources, the model will focus on enhancing realism, producing images that closely resemble authentic photography. This shift in emphasis is seen as a response to ongoing demands for higher-quality outputs in the AI imaging sector.
Additionally, gpt-image-2 is expected to improve upon the composition of generated visuals, allowing for more accurate placement of text, objects, and overall layouts. These enhancements could make the model particularly useful for practical applications such as presentations, design mock-ups, and visual storytelling, potentially expanding its appeal beyond casual users to professionals in various fields.
In parallel, ChatGPT is testing a new reference photo feature that will allow users to upload an image once and reuse it across multiple prompts. This feature aims to streamline workflows, reducing the need for repetitive uploads and enhancing user experience.
If the reports regarding gpt-image-2 prove accurate, OpenAI’s new offering will directly compete with advanced image generation tools from Google and Anthropic. However, significant details about the model, including its release timeline and specific benefits, remain undisclosed.
OpenAI is also reportedly working on a new version of its language model, GPT 5.5, codenamed Spud, which is anticipated to launch next week. This model is expected to introduce notable enhancements over existing iterations, although official confirmation is still pending.
The race for superior AI-generated imagery is intensifying, with various players in the sector aiming to capture market share by providing tools that meet evolving user demands. As OpenAI continues to refine its offerings, the technology landscape may witness significant shifts, particularly in the realms of creative and practical applications.
See also
Veo 4 Video Generator Launches, Enabling Instant AI Video Creation with Simple Prompts
AI-Driven AdTech Surpasses $800 Billion as Platforms Optimize User Journeys
Deep Generative Models Enhance Engineering Design Efficiency in BMW A-Pillar Project
OpenAI Launches ChatGPT Images 2.0, Surpassing Competitors in Image Generation Quality
OpenAI Unveils ChatGPT Images 2.0, Enhancing Text Accuracy and Resolution Up to 2K
