Driven by advances in multimedia technologies, digital multimodal composing (DMC) has gained growing scholarly interests. DMC affords language learners a new experience in language learning and use. While completing DMC projects, L2 learners construct meanings by assembling both linguistic and non-linguistic resources (e.g., visuals, audio, videos, etc.). This is largely in line with their out-of-class text composing experiences. To understand the state-of-the-art developments in DMC research, this article reviewed the empirical studies published between January 2011 and October 2022. It provides a general overview of these studies regarding the overall trends, theoretical underpinnings, research contexts, research methods, and major research strands. The research findings indicate: (1) DMC has been increasingly attended to by L2 researchers with the number of research articles showing an ascending trend; (2) These empirical studies were grounded in a variety of theories, including the social semiotic theory of multimodality, multiliteracies, task-based language teaching, social/cultural construction, meta-functions of systemic-functional linguistics, translanguaging, and others; (3) The bulk of research has been investigated in tertiary schools, although the research contexts were diversified; (4) Most studies were conducted qualitatively; (5) Scholarly attention has been paid to four major strands: composing process, teachers' and learners' perceptions of DMC, DMC's impact on L2 development, and assessment of DMC artifacts. The major research strands and sample studies were discussed in detail. This article ends with recommendations for future studies.