Meta AI has unveiled Muse Image and Muse Video, its initial multimedia generation models from Meta Superintelligence Labs. Muse Image is described as Meta's most advanced image model, capable of precise editing, multi-reference composition, and incorporating social context from Instagram. It also features agentic tool use and integrates with Muse Spark. Muse Video, currently in preview, shares a similar pretraining base with Muse Image and includes native audio support. Both models are being rolled out in the Meta AI app, web interface, and integrated into Instagram Stories and WhatsApp in select regions. AI
IMPACT These models could enhance content creation capabilities across Meta's platforms, potentially influencing user engagement and creator tools.
RANK_REASON New model release from a frontier lab (Meta AI). [lever_c_demoted from frontier_release: ic=2 ai=1.0]
Read on Mastodon — fosstodon.org →
AI-generated summary · Google Gemini · from 2 sources. How we write summaries →
