XiaoYing Tech CEO Han Sheng is confirmed to speak at QCon Shanghai 2026 on the three paradigm shifts in content creation: from tools to AIGC to Agents. The preview highlights that the core challenge has moved from single-model capabilities to enabling AI to complete end-to-end complex tasks, focusing on engineering issues like requirement understanding, multi-model orchestration, and scalable delivery. Specific technical details or benchmark data remain undisclosed.
Multi-model Orchestration
多模型编排
A technical architecture coordinating multiple distinct AI models to work together on complex tasks.
Curated Intelligence
Google Cloud unveiled Gemini Agent at the Gemini at Work 2026 event, introducing the concept of 'Coworker Agents' that possess independent enterprise identities including email, calendar, and persistent storage. The agent executes cross-application tasks via a single prompt, supports sub-agent collaboration, and breaks model lock-in by currently integrating Anthropic Claude with plans for broader multi-model support. It is currently in private preview for enterprise knowledge work scenarios.
Google announced the upgrade of Gemini into a unified agentic AI capable of autonomous task execution, initially targeting enterprise users. Beyond answering questions, the agent can perform complex tasks such as code generation, meeting scheduling, and travel booking on behalf of users. Additionally, Google introduced new features including multi-model orchestration, smart routing, and real-time spend caps to help businesses better control their enterprise AI costs during large-scale deployment.