JD.com has open-sourced its interactive audiovisual world model, JoyAI-EchoWM, along with EchoWM-Flash. The model achieved scores of approximately 81.6-81.7 on the WBench Navigation benchmark, demonstrating unified camera-intent control and native audio-video generation capabilities. JD.com also announced plans to develop a large domestic GPU cluster and its Superbrain 3.0 logistics system. AI
IMPACT This release contributes to the open-source community's multimodal AI capabilities, potentially enabling new applications in audio-visual generation and navigation.
RANK_REASON Open-source release of a multimodal model with benchmark scores. [lever_c_demoted from research: ic=1 ai=1.0]
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →