Tencent Hunyuan Hy4 Preview Lite Shrinks Model Weights to 214GB
Tencent's Hunyuan team released the Hy4 Preview Lite on September 1, using two core technologies to compress model weights from nearly 1.5TB to about 214GB. The compressed model matches the original on long-text comprehension and multi-turn long-context retrieval, with only a slight dip in math ability, covering daily coding assistance, tool calling, long-document processing, and general Q&A.
Tencent's Hunyuan team announced the release of the Hy4 Preview Lite on September 1. Using two core technologies, the team compressed the model's weights from nearly 1.5TB to approximately 214GB.
On mainstream tasks, the compressed model's long-text comprehension is nearly on par with the original, multi-turn long-context retrieval is essentially at the same level, and math ability shows only a slight decline. The model is suited for everyday coding assistance, tool calling, long-document processing, and general question-answering scenarios.
Why this event matters
The event has a measured impact on 1 industry. The strongest current signal is positive for Artificial Intelligence, with intensity 70/100 and 80% confidence over a short term horizon.
Impact figures are analytical estimates that combine direction, intensity, confidence and event importance. They are not investment advice.