Background　Extreme　wildfires　pose　a　serious　threat　to　forest　vegetation　and　human　life　because　they　spread　more　rapidly　and　are　more　intense　than　conventional　wildfires.　Detecting　extreme　wildfires　is　challenging　due　to　their　visual　similarities　to　traditional　fires,　and　existing　models　primarily　detect　the　presence　or　absence　of　fires　without　focusing　on　distinguishing　extreme　wildfires　and　providing　warnings.Aims　To　test　a　system　for　real　time　detection　of　four　extreme　wildfires.Methods　We　proposed　a　novel　lightweight　model,　called　LEF-YOLO,　based　on　the　YOLOv5　framework.　To　make　the　model　lightweight,　we　introduce　the　bottleneck　structure　of　MobileNetv3　and　use　depthwise　separable　convolution　instead　of　conventional　convolution.　To　improve　the　model＇s　detection　accuracy,　we　apply　a　multiscale　feature　fusion　strategy　and　use　a　Coordinate　Attention　and　Spatial　Pyramid　Pooling-Fast　block　to　enhance　feature　extraction.Key　results　The　LEF-YOLO　model　outperformed　the　comparison　model　on　the　extreme　wildfire　dataset　we　constructed,　with　our　model　having　excellent　performance　of　2.7　GFLOPs,　61　FPS　and　87.9%　mAP.Conclusions　The　detection　speed　and　accuracy　of　LEF-YOLO　can　be　utilised　for　the　real-time　detection　of　four　extreme　wildfires　in　forest　fire　scenes.Implications　The　system　can　facilitate　fire　control　decision-making　and　foster　the　intersection　between　fire　science　and　computer　science.　We　tested　a　lightweight　architecture　called　LEF-YOLO　for　detecting　four　extreme　wildfires.　We　found　improved　detection　accuracy　through　multi-scale　fusion　and　attention　mechanism,　and　constructed　four　extreme　wildfire　datasets　and　compared　these　with　multiple　object　detection　models　and　lightweight　feature　extraction　networks.　This　method　is　beneficial　for　the　development　of　extreme　wildfire　field　robots.