This paper presents a fire and smoke detection system using fine-tuned YOLOv8 and YOLOv7 models. The study aims to improve the accuracy and efficiency of detecting fire and smoke in various environments. A dataset of over 11,000 images was used for training and testing. The YOLOv8 model achieved a mAP:50 of 92.6%, a precision score of 83.7%, and a recall of 95.2%. The results were compared with YOLOv6, Faster-RCNN, and DETR models. The YOLOv8 model outperformed these models in terms of precision, recall, and mAP:50. The study also evaluated the performance of different YOLOv8 versions, including YOLOv8n, YOLOv8s, YOLOv8m, YOLOv8l, and YOLOv8x. The YOLOv8x model achieved the highest precision, recall, and mAP:50 scores. The results show that the YOLOv8 model is effective for fire and smoke detection in various environments, including forests, streets, and industrial zones. The study also highlights the limitations of the model, such as sensitivity to weather conditions like haze, clouds, and fog. The authors suggest that increasing the size of the training dataset and extracting deeper features can improve the model's performance. The proposed model can be deployed in various industrial applications for fire and smoke detection. The study concludes that the YOLOv8 model is a promising solution for fire and smoke detection due to its high accuracy, efficiency, and adaptability to different environments.This paper presents a fire and smoke detection system using fine-tuned YOLOv8 and YOLOv7 models. The study aims to improve the accuracy and efficiency of detecting fire and smoke in various environments. A dataset of over 11,000 images was used for training and testing. The YOLOv8 model achieved a mAP:50 of 92.6%, a precision score of 83.7%, and a recall of 95.2%. The results were compared with YOLOv6, Faster-RCNN, and DETR models. The YOLOv8 model outperformed these models in terms of precision, recall, and mAP:50. The study also evaluated the performance of different YOLOv8 versions, including YOLOv8n, YOLOv8s, YOLOv8m, YOLOv8l, and YOLOv8x. The YOLOv8x model achieved the highest precision, recall, and mAP:50 scores. The results show that the YOLOv8 model is effective for fire and smoke detection in various environments, including forests, streets, and industrial zones. The study also highlights the limitations of the model, such as sensitivity to weather conditions like haze, clouds, and fog. The authors suggest that increasing the size of the training dataset and extracting deeper features can improve the model's performance. The proposed model can be deployed in various industrial applications for fire and smoke detection. The study concludes that the YOLOv8 model is a promising solution for fire and smoke detection due to its high accuracy, efficiency, and adaptability to different environments.