Since February 2022, the war in Ukraine has been a staple of the European media coverage. While the public across the continent staunchly rooted for Ukraine in the earlier stages of the war, support is cooling off in various European countries. Considering the increasing role social media plays in informing citizens, we ask the following research questions: (1) what information do European publics receive about the war in Ukraine from mainstream media via social media, and what overlap, if at all, is there between the visual and verbal coverage in different countries? (2) How is the war covered, episodically or thematically? (3) How are visuals related to the text – providing supporting information, extending the text meaning, or contrasting the text? We collected about 180K Facebook posts from 48 top news organizations across 24 European countries, over two months, corresponding to the two-year anniversary of the Russian attack (February-March 2024), and to new developments on the Eastern Front (May-June 2024). We use the computer vision capabilities offered by ammico (github.com/ssciwr/AMMICO, Dumitrescu et al., 2023) to analyze the images used in the media posts (e.g., color composition and object detection), and extract the text. We then investigate the episodic/thematic aspect of the coverage (by looking at visual object depiction and text topic) and examine the emotional content in visuals and text (through object depiction and text sentiment analysis). The analysis provides unique insights into the European public’s window on the conflict and highlights the computer vision capabilities for media coverage analysis.