Newshot

Tencent Hunyuan Hy4 flash model released: 224B total parameters, 12B activated, cost per task about 5% of Hy4

AI summary

The Tencent Hunyuan team officially launched Hy4 flash, positioning it as the most cost-effective Hy model to date. It has 224B total parameters, including 64B Engram parameters, activates 12B parameters per token, supports multimodal input and a 1M token context length, and its API price is substantially lower than Hy4. Tencent says the cost-effectiveness comes from coordinated optimization across pre-training, post-training, and inference, and that on its WorkIndex benchmark the cost per task is only 1/13 of Hy4 preview, 1/20 of Hy4, and 1/8 of DeepSeek-V4.1-Flash-1017 (max tier, peak-hour pricing).

Translation(model translation, not the source's own text)

Tencent Hunyuan Hy4 flash Model Released: 224B Total Parameters, 12B Activated, Cost per Task About 5% of Hy4

IT之家 November 16 news, the Tencent Hunyuan team today announced the official launch of Hy4 flash, which it positions as the most cost-effective Hy model to date.

The model has 224B total parameters, including 64B Engram parameters, activates 12B parameters per token, supports multimodal input and a 1 million token context length, and its API price is significantly lower than Hy4.

Tencent said Hy4 flash's high cost-effectiveness comes from coordinated optimization across three stages: pre-training, post-training, and inference. In the pre-training stage, architectures such as hybrid attention, LatentMoE, and Engram (IT之家 note: Engram is a scalable lookup-table-style parameter module used to expand model capacity at lower compute cost) were introduced, reducing KV cache size to 1/24 of Hy3 and 1/6 of Hy4. In the post-training stage, comprehensive intelligence is obtained through self-distillation from Hy4, and then reinforcement learning compresses end-to-end task output length by 50%. In the inference stage, methods including megakernel, 4bit quantization, and dflash are combined to achieve decoding speeds exceeding 150 tps and support large-scale low-cost deployment and operation.

On WorkIndex, the official evaluation set of typical WorkBuddy tasks, the cost per task of Hy4 flash is only 1/13 of Hy4 preview, 1/20 of Hy4, and 1/8 of DeepSeek-V4.1-Flash-1017 (max tier, busy-hour pricing).

In terms of overall results, Hy4 flash is slightly better than Hy4 preview. This comparison framework focuses on the output quality of everyday work, single-turn output length, and end-to-end time, rather than public leaderboard scores.

To measure real-world user experience, Tencent invited 85 internal WorkBuddy users to conduct blind tests on 259 everyday work tasks across 27 categories.

The results show that Hy4 flash's user satisfaction is close to Claude Haiku 5.5, its cost per task is significantly lower than all models in the comparison, and its task duration is also the shortest.

Hy4 flash has already launched on products including Tencent Cloud TokenHub, WorkBuddy, Yuanbao, WeCom, QQ, ima, and Marvis, and will soon be integrated into OpenRouter.

Before November 30, 2026, users can use it for free for a limited time on WorkBuddy, and Tencent Cloud TokenHub API is available at a limited-time 50% discount.

Tencent recommends using Hy4 flash for everyday tasks such as deep search, document processing, PPT generation, simple front-end display, and Computer Use, while complex tasks such as long-horizon software engineering should be left to Hy4, which has a higher intelligence ceiling. The company also said it plans to open-source Hy4 flash and maintain high-frequency iteration.

Related Articles

  • Tencent's General Office calls for starting anew in the AI era, official original text released
  • Tencent Hunyuan releases real-time speech model Hy Realtime: can speak, sing, and call tools, already launched in Yuanbao voice calls, Tencent customer service, and more
  • Tencent Hunyuan: Hy4 calls surge, WorkBuddy has expanded tenfold, Hy3 free period extended
  • Tencent Hunyuan Hy4 official version released, claimed to rank among the strongest open-source models for the first time and built for real productivity
  • Tencent Hunyuan releases Hyra 2.0: building reusable tools for model R&D
  • First introduction of visual understanding, Tencent Hunyuan Hy4 and Hy4 flash AI models to be released soon

The translation is machine-generated and individual terms may be off. Nothing is rewritten or cut; the original is authoritative.

Published by IT之家
This site only summarises and aggregates; copyright stays with the original author. Headlines and summaries follow the source — where they differ, the original wins.
Read the original ↗