The OpenAI company has taken a step in the development of artificial intelligence and opened access to its multimodal GPT-IMAGE-1 via API. The solution based on the GPT-4O architecture allows you to integrate the generation of images into third-party services, who has already attracted the attention of large market players, including Adobe, Figma and Canva.
According to the developer, the model can create images in different styles, accurately reproduce textements and adapt to user requirements. This opens opportunities for use in design, E -commerce, education and other areas. Figma tests for example the image editing function for Tekstquerys and Adobe plans to add support for the model in Firefly and to express them for experiments with styles.
Less obvious market participants also show interest in technology. So the Instacart service studies the possibility to generate illustrations for recipes, and the Opusclip platform uses AI to make a video on YouTube. OpenAI also noted that Chatgpt users generated more than 700 million images in the first week after the release in March 2025, which temporarily overloads the system.
To meet ethical standards, filters are introduced in the API, which blocks the creation of malignant content and all images contain Metadata C2PA to verify origin. According to the company, the data from API users is not used to train models and developers can configure the level of moderation of content. Currently, GPT-IMAGE-1 is available through images API and support in reactions API is expected in the near future.
🇷🇺 For Remaster Tes 4: Oblivion, a high -quality Russifier came out – the one who was in the original (good)
Source: VG Times
Gregory Robert is a sports aficionado and a writer for “Social Bites”. He provides in-depth coverage of the latest sporting events and trends, offering a unique and knowledgeable perspective on the world of sports.