Tencent releases Hy4 flash, a 224B-parameter model with 12B active parameters

Nov 17, 2026
Tencent says the model matches its larger Hy4 Preview at lower cost and supports a 1M-token context; it is free on WorkBuddy until November 30.

Tencent has released Hy4 flash, a new model in its Hy family with 224 billion total parameters, including 64 billion Engram parameters, but only 12 billion active parameters, the company said in a press release. It accepts multimodal inputs and supports a context of up to 1 million tokens.

Tencent says Hy4 flash slightly outperforms the much larger Hy4 Preview, which has 770 billion total and 49 billion active parameters, at significantly lower cost on its internal WorkIndex benchmark, which covers tasks such as editing slide decks and documents and repairing spreadsheets. In a blind test with 85 internal users across 259 tasks, it reached user satisfaction close to Claude Haiku 5.5, with the lowest per-task cost of the models evaluated.

The model is available on Tencent Cloud TokenHub and in Tencent products including WorkBuddy, Yuanbao, WeCom and QQ, with OpenRouter integration coming soon. It is free on WorkBuddy until November 30 and half price on TokenHub. Tencent recommends it for everyday work such as deep research, document processing and computer use, and the full Hy4, released on November 4, for more complex tasks.

We hope you enjoyed this article

Consider subscribing to one of our newsletters like Daily AI Brief.

Also, consider following us on social media:

Free newsletter

Daily AI Brief

Daily report covering major AI developments and industry news, with both top stories and complete market updates

Whitepaper

Tensordyne Napier: What If One Rack Could Do the Work of Nine?

Tensordyne

This Tensordyne whitepaper presents Napier, an inference-focused AI processor and rack-scale system based on the company’s TDN Math logarithmic number system. It examines infrastructure requirements for large mixture-of-experts and agentic models, compares major inference architecture approaches, and details the TDN AIP processor, TDN72 pod, TDN Link fabric, and Napier Ultra configuration. The paper reports simulation-based performance, cost, and accuracy-validation results, including Tensordyne’s projected comparison of one Napier rack with a nine-rack Nvidia Rubin plus Groq deployment; the chip is reported as taped out and in fabrication.

Read more
Free, six days a week

Daily AI Brief: the AI news that matters, in your inbox.