Tencent released the weights of its flagship Hy4 AI model on Friday, allowing developers to download and deploy the model themselves. The release follows Hy4’s initial launch on November 4 and makes the flagship available under the Apache 2.0 license, which permits commercial use and modification.

Hy4 supports text, image and video understanding, with a context window of about one million tokens. Its mixture-of-experts language backbone contains 770 billion parameters, with 49 billion activated for each token. Tencent says the model improves on Hy4 preview in instruction following, long-context understanding, extended tasks and research.

The model supports workflows spanning graphical interfaces and command-line tools, alongside frontend development, game development and office tasks. Tencent provides deployment instructions for vLLM and SGLang, as well as fine-tuning resources and a safety evaluation card. The flagship’s release follows the separate opening of the smaller Hy4 flash model. [Tencent]