# Tencent's Hy4 preview expands open-weight model scale and context

_Saturday, August 29, 2026 at 7:53 PM EDT · AI, Infrastructure · Latest · Tier 2 — Notable_

![Tencent's Hy4 preview expands open-weight model scale and context — Primary](https://static.simonwillison.net/static/2026-08-29/IMG_7725.jpeg)

Tencent introduced Hy4 Preview, an open-weight text-input language model with 770 billion total parameters, 49 billion active parameters and a one-million-token context window. The model files occupy 1.56TB on Hugging Face, according to the source. Its predecessor, Hy3, released in July, had 295 billion total parameters, 21 billion active parameters, a 256,000-token context window and a 598GB footprint. Hy4's chat template exposes high and no_think reasoning settings.

## Sources

- [Simon Willison's Weblog](https://simonwillison.net/2026/Aug/29/hy4/)

---
Canonical: https://techandbusiness.org/newswire/o-P2tZ9QaRPnPdfV-C3QH1
Retrieved: 2026-08-30T02:48:15.543Z
Publisher: Tech & Business (techandbusiness.org)