What drives variation in media framing of international conflicts, and how do the media shape public attitudes? We examine media slant in the conflict in Gaza from October 7, 2023, through late 2024. We collect the headlines, article text, and images in every article mentioning the conflict in the 20 largest newspapers in the US. We code the articles to measure features of the coverage including: mentions of civilians, harms, and levels of humanization. Using word embeddings, we describe differences in how Israelis and Palestinians are covered, documenting variation over time and across news outlets. We then investigate whether differences in coverage across outlets are primarily supply- or demand-driven. To identify supply and demand factors, we exploit common ownership of news outlets, geographic variation in audiences, and exogenous changes in reliance on news wire services. Finally, we connect media coverage to public opinion toward the conflict using an original survey experiment. This research advances our understanding of how media coverage shapes public perceptions of identity-based conflicts and humanitarian crises in the Middle East.