Today, we’re excited to introduce CM3leon (pronounced like 'chameleon'), a versatile foundation model that seamlessly handles both text-to-image and image-to-text generation.