OpenAI releases' new flagship image generation model 'GPT Image 1.5

苦难深重难言 · 发表于 2025-12-17 14:29:05

OpenAI has comprehensively upgraded its image generator, launching a new model and redesigned interface. On December 16th, the company released its "new flagship image generation model" GPT Image 1.5, stating that it can better follow instructions, edit photos in specific ways, and generate results up to four times faster. The model is now open to all users.

The company stated in a press release on its official website that the new model will be able to better "fit your intentions" by modifying photos, including "more practical photo editing functions, more realistic clothing and hairstyle fitting effects, as well as style filters and conceptual transformations that can preserve the essence of the original image. The newly added "Images" tab in the ChatGPT sidebar will include preset filters, popular tips, and creative ideas.

After the viral success of Google's Nano Banana, the competition for image generation among artificial intelligence labs has become increasingly fierce.

OpenAI claims that GPT Image 1.5 marks a shift from novel image generation to practical, high fidelity visual creation, and transforms ChatGPT into a fast and flexible creative studio for daily editing, expressive transformations, and practical applications.

[AI] OpenAI releases' new flagship image generation model 'GPT Image 1.5

回复