OpenAI has launched an upgraded version of ChatGPT Images featuring GPT Image 1.5, their latest flagship image generation model. The enhancement delivers outputs up to 4x faster while providing more precise editing capabilities and improved consistency in preserving details across image modifications.
Key Improvements in Precision and Control
The updated model excels at maintaining essential elements like facial features, lighting, and composition while making requested changes. This capability enables more reliable photo editing, realistic virtual try-ons for clothing and hairstyles, and creative filters that preserve the original image's essence. The model demonstrates stronger instruction adherence, allowing for intricate compositions where element relationships remain as intended.
Enhanced Text Rendering and Quality
GPT Image 1.5 shows significant progress in text rendering, now handling denser and smaller text more effectively. The model produces more natural-looking outputs with improved rendering of multiple small faces and better overall image quality. These improvements translate to immediately usable outputs across various use cases.
New Creative Interface and Workflow
OpenAI has introduced a dedicated Images section within ChatGPT, accessible through the sidebar on both mobile and web platforms. This space includes dozens of preset filters and trending prompts that update regularly to reflect emerging trends. Users can save their likeness for reuse across future creations without repeatedly accessing their camera roll. The interface allows continuous generation of new images while others are still processing, enabling faster exploration of creative ideas.
API Implementation and Business Applications
GPT Image 1.5 is available through OpenAI's API with all the same improvements as the ChatGPT version. The model offers better preservation of branded logos and key visuals across edits, making it particularly suitable for marketing, brand work, and e-commerce applications. API pricing has been reduced by 20% compared to GPT Image 1.0, allowing developers to generate more images within the same budget.
Creative Transformation Capabilities
The model demonstrates enhanced creativity through transformations that modify and add elements while preserving important details. These transformations work effectively for both simple adjustments and complex concepts, accessible through preset styles in the new ChatGPT Images feature without requiring written prompts.
Current Limitations and Future Development
While GPT Image 1.5 represents meaningful progress, OpenAI acknowledges that results remain imperfect with significant room for improvement in future iterations. The model shows clear improvements across various use cases compared to the initial version, though some limitations persist in areas like multiple face rendering and multilingual text generation.
Availability and Access
The new ChatGPT Images model is rolling out to all ChatGPT users globally and is available in the API as GPT Image 1.5. The feature works across models without requiring specific selection. The previous version of ChatGPT Images remains accessible as a custom GPT for users who prefer the earlier implementation.