ZeroSearch: Incentivize the Search Capability of LLMs without Searching Paper β’ 2505.04588 β’ Published 1 day ago β’ 39
HunyuanCustom: A Multimodal-Driven Architecture for Customized Video Generation Paper β’ 2505.04512 β’ Published 1 day ago β’ 20
FramePack Video Gen Collection fast & compact video generation with FramePack - a next-frame prediction neural network structure that generates videos progressively β’ 7 items β’ Updated 2 days ago β’ 2
view post Post 3619 ACE-Step π΅ a music generation foundation model released by StepFun & ACEStudio Model: ACE-Step/ACE-Step-v1-3.5BDemo: ACE-Step/ACE-Stepβ¨ 3.5B, Apache2.0 licensedβ¨ 115Γ faster than LLMs (4-min music in 20s on A100)β¨ Diffusion + DCAE + linear transformer = speed + coherenceβ¨ Supports voice cloning, remixing, lyric editing & more See translation 1 reply Β· π 7 7 + Reply