Alibaba Launches Qwen Create: Wan3.0 Video Model & Multi-Agent Teams
机器之心 · wechat · 2026-08-31
Alibaba launched "Qwen Create", an AI platform integrating the Wan3.0 video generation model and the AgentTeams multi-agent workflow. Wan3.0 supports 30-second single-take generation, multi-modal referencing, and pixel-level control, with pricing as low as 0.26 RMB per second for 720P. The AgentTeams feature allows users to coordinate agents acting as directors, screenwriters, and artists via natural language, streamlining the pipeline from script to final cut and solving issues of cross-tool asset transfer and consistency.
More from Multimodal
- Complex AI video prompt visualizing conversation as mechanical printing architecture — umesh_ai · 2026-08-31
- Open-Source AI Tools: Run Stable Diffusion Models via WebUI Locally — goyalshaliniuk · 2026-08-31
- ComfyUI: Build Advanced Node-Based Image-Generation Workflows — goyalshaliniuk · 2026-08-31
- LayerRecall: layer-wise memory routing for long-horizon consistency in video generation — zju · 2026-08-31
- Best settings for training LoRAs on ANIMA models? — XZtext18 · 2026-08-31
- Prompt: Creating a 30-second cinematic horror transformation — LudovicCreator · 2026-08-31