TMTPOST — Alibaba’s Tongyi QWen has unveiled Qwen VLo, a new multimodal model designed for unified understanding and generation across text and image inputs. The model introduces a novel generative mechanism that produces content gradually—from top to bottom and left to right—making it particularly effective for generating long-form text with fine-grained control. Qwen VLo is now available for public use on the Qwen Chat platform (chat.qwen.ai).