Create and transform images with unprecedented context awareness. Use both text and images as prompts to seamlessly extract and modify visual concepts with up to 8x faster speed than leading models.