OpenAI has launched GPT Image 1.5 for ChatGPT, marking a significant upgrade in its image generation capabilities. The new model boasts a fourfold increase in generation speed and enhanced instruction following. It also delivers more precise editing, maintaining consistency in elements like lighting and appearance across prompts and edits.
GPT Image 1.5 is available to all ChatGPT users and via the API as GPT Image 1.5. The upgrade includes a dedicated space within ChatGPT for image creation, featuring preset styles and trending designs. For developers, the model offers a 20% reduction in image input and output costs compared to its predecessor. OpenAI's advancements intensify the competition with Google's Gemini in the AI image generation arena.
The model excels at various editing functions, including adding, subtracting, combining, and transposing image elements. It also demonstrates improved text rendering in generated images, even with denser and smaller text. Benchmarks show that GPT Image 1.5 has already surpassed Google's Nano Banana Pro in text-to-image performance.
