Recent advancements in Generative AI techniques have transformed countless real-world applications, transforming intelligent human-like conversational content creation with Large Language Models such as GPT. Furthermore, generative adversarial networks and Transformers have revolutionized picture, music, and movie making and have provided topic-based idea recommendations. When models were combined with vision approaches, this resulted in extremely Large Vision-Language Models, which considerably improved their comprehension of the contents of imagery. In summary, unlike the standard Large Vision-Language Models, our model, Spectate-GPT, goes a step further in attempting to mimic human-like help by integrating common-sense reasoning. Spectate-GPT uses the provided images to analyze and produce summaries while responding to inquiries similar to a chatbot. By fine-tuning the Mixture of Experts to deal with the performance and computation issues, the modeled greater Vision-Language Model is well precise and economical. From the comparison analysis it is observed that the proposed approach, i.e., Spectate-GPT is performing better as compared to the earlier approaches.