GPT Image 1.5 is OpenAI's most advanced text-to-image and image-to-image model, released in December 2025. It processes images and text in a unified neural network, generating high-fidelity images up to 4x faster than its predecessor while maintaining exceptional precision in edits. The model excels at preserving composition, lighting, and fine-grained details across iterative edits, handles complex text rendering within images, and supports creative transformations including style transfer, temporal shifts, and conceptual combinations. Key capabilities include precise image editing that maintains facial likeness and lighting consistency, superior text rendering for signage and layouts, and the ability to follow detailed instructions reliably.