HiDream.ai has launched its first native multimodal video generation model, HiDream-O1-Video-1.0 (HD-V1), capable of producing 5-20 second 1080p videos from various inputs including text, images, and video. The model reportedly excels in understanding intent, adhering to physical laws, and generating coherent narratives with synchronized audio and video. HD-V1 has achieved top rankings on the Artificial Analysis Image to Video Leaderboard and Arena.ai Image-to-Video benchmarks, positioning HiDream.ai as a leading Chinese force in the competitive AI video generation landscape. AI
IMPACT Sets a new benchmark for multimodal AI video generation, potentially accelerating competition in the field.
RANK_REASON Frontier-lab model release with system card [lever_c_demoted from frontier_release: ic=1 ai=1.0]
- Arena.ai Image-to-Video
- Artificial Analysis Image to Video Leaderboard
- HD-V1
- HiDream.ai
- HiDream-O1
- HiDream-O1-Image
- HiDream-O1-Video-1.0
- HiDream-O1-World
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →