Text-to-comic generation has emerged as a captivating domain within artificial intelligence, into the refinement of prompt training for fine tuning of LLM model blending textual narratives with visual storytelling to meet the rising demand for personalized and interactive media experiences. This study delves for text-to-comic generation, specifically focusing on fine-tuning the Mistral 7B language model. By harnessing Mistral 7B’s advanced capabilities, the aim is to enhance the generation of prompts tailored for comic creation, addressing challenges in maintaining the narrative coherence, achieving accurate text-image alignment and capturing character essence. The methodology encompasses data pre-processing, model fine-tuning, and comprehensive evaluation. The Mistral 7B model was fine-tuned on the curated Alpaca dataset using the Low-Rank Adaptation (LoRA) technique, enabling efficient specialization for comic generation tasks. The evaluation process, utilizing ROUGE metrics, revealed substantial improvements in recall, precision, and F1 scores compared to the base Mistral 7B model. These results demonstrate that the fine-tuned model significantly enhances the quality of comic prompts, producing more coherent storylines and visually aligned dialogues, thereby advancing the text-to-comic generation pipeline.