mixture-of-experts — A model built from many specialist sub-networks where only a few activate per token, giving big-model capability at small-model running cost.
tool use — A model's ability to call external functions — run code, search the web, edit files — instead of only generating text.
AI agent — An AI system that doesn't just answer once but works toward a goal in a loop — taking actions, reading the results, and deciding what to do next.
context window — The maximum amount of text a model can consider at once — its working memory for the current conversation or task.
Why it matters
Hy3 preview is open weights: 295B total with 21B active, 256K context windowThe maximum amount of text a model can consider at once — its working memory for the current conversation or task.Full definition →, on Hugging Face and GitHub with vLLM and SGLang support, priced from 1.2 yuan per million input tokens and free on OpenRouter for two weeks. It plugs into OpenClaw, OpenCode and KiloCode.
articleTencent Launches Qclaw Globally Lowering Barriers To Ai Agent Deployment
article腾讯推出Qclaw海外版，降低Ai智能体部署门槛
articleTencent Unveils Hy3 Preview Model Enhances Agent Capabilities And Real World Usability
articleTencent Hunyuan Officially Releases Hy3 Advancing Agent Capabilities And Deeper Product IntegrationTencent