Tencent Releases Hy4 Flash AI Model
Tencent Hy4 flash offers multimodal AI capabilities, up to 1M tokens of context and lower-cost deployment for workplace tasks.

Tencent has released Hy4 flash, a multimodal AI model designed for everyday productivity workflows. The model has 224 billion total parameters, including 64 billion Engram parameters, with only 12 billion active parameters and a context length of up to 1M tokens. Tencent prices Hy4 flash at approximately US$0.070 per million input tokens, US$0.281 per million output tokens and US$0.003 per million cached input tokens. It is now available through Tencent Cloud TokenHub, WorkBuddy, Yuanbao, WeCom, QQ, ima and Marvis.
Engineering For Efficiency
Hy4 flash uses hybrid attention, LatentMoE and Engram architectural features in its pre-training pipeline. Its KV cache size is just 1/24 that of Hy3 and 1/6 that of Hy4, while reinforcement learning cuts average end-to-end task output tokens by 50 percent.
Combining optimization using megakernel, 4-bit quantization, and DFly techniques, Hy4 flash achieves decoding speeds exceeding 150 tokens per second, enabling low-cost deployment at scale.
Workplace Task Performance
The company’s internal WorkIndex benchmark assesses PowerPoint deck creation and editing, document editing, spreadsheet repair, audio/video processing and visual understanding tasks typical of use with WorkBuddy. Hy4 flash slightly outperforms the larger Hy4 Preview (770 billion total parameters, 49 billion active parameters) in overall effectiveness, at a significantly lower cost.
Based on WorkIndex, Hy4 flash delivers a per-task cost of just 1/13 that of Hy4 Preview, 1/20 that of Hy4, and 1/8 that of DeepSeek-V4.1-Flash-1017 (Max tier, peak-hour pricing). In a blind evaluation with 85 internal Tencent WorkBuddy users across 259 tasks in 27 categories, it achieved user satisfaction close to Claude Haiku 5.5, with the significantly lower per-task cost among all models evaluated and the shortest task completion time.
Availability And Planned Updates
From now till November 30, users can access Hy4 flash for free on WorkBuddy and get 50 percent off standard pricing via Tencent Cloud TokenHub. OpenRouter integration is coming soon.
The team recommends Hy4 flash for most everyday tasks, including deep research, document processing, PowerPoint deck generation, simple front-end displays, and computer use, while directing users to the more capable Hy4 for complex tasks such as long-horizon software engineering. Tencent Hy also plans to open-source Hy4 flash with frequent iterations to continuously improve its practical utility and efficiency for real-world work.



