The development of Multimodal Large Language Models (MLLMs) is evolving rapidly, enabling neural networks to receive and process visual, audio, and other types of data, making them even more human-like and intelligent. MLLMs can mimic human cognitive processes, expanding the application of AI in practical settings, such as creating captions based on images, making medical diagnostics based on visual data, developing assistive technologies for impaired individuals, optimizing UI of websites and coding, and solving math problems based on diagrams, graphs, and charts. The lineup of available MLLMs is constantly expanding, with popular models including GPT-4V, LLaVa, and DALL-E, each offering unique capabilities and limitations. Research efforts are ongoing to enhance MLLM performance, address vulnerabilities, and develop new techniques for multimodal learning, such as Multimodal Instruction Tuning (M-IT), Multimodal In-Context Learning (M-ICL), and LLM-Aided Visual Reasoning (LAVR). As MLLMs become more widespread, tools like VectorShift are emerging to facilitate the introduction of multimodal AI into applications with intuitive no-code functionality and SDK interfaces.