Hy3 Preview

Hy4 Preview is Tencent Hunyuan’s large language model for agents, coding, office automation, and complex tool use.Hy4 preview has a total of 770B parameters and 49B active parameters, and is primarily optimized for agent, coding, and productivity scenarios. It strengthens understanding, planning, tool invocation, and sustained execution capabilities for complex tasks. Compared with the previous generation, Hy4 preview further improves multi-step agents, code development, and productivity tasks, offering better task decomposition, context continuity, instruction following, and long-horizon execution. In coding scenarios, it further enhances code understanding, generation, modification, and the handling of complex engineering tasks; in productivity scenarios, it focuses on improving document processing, information analysis, office automation, game development, webpage generation, and cross-tool collaboration. Hy4 preview is suitable for coding agents, complex tool invocation, and various agent workflows that require multi-step planning and sustained execution, providing more reliable task completion for complex real-world business scenarios.

The Hy3 official version is honed for real-world business scenarios, using a Mixture-of-Experts (MoE) architecture with 295B total parameters and 21B activated parameters. It natively supports a 256K context window and offers multiple thinking modes: no_think (ultra-fast response), think_low (quick thinking), and think_high (deep reasoning), balancing ultra-fast responses, complex reasoning, and invocation cost. Compared with the Preview version, Hy3—based on real business feedback from Tencent Yuanbao, WorkBuddy, ima, Marvis, and others—focuses on improving the Coding Agent, long-form understanding, multi-turn context continuity, search QA, and complex task execution, performing more stably in reducing hallucinations, improving task completion, and engineering usability. It is better suited to practical scenarios such as frontend tasks, cross-file code development, long-document analysis, office automation, and multi-step Agent workflows.

The Hy3 official version is honed for real-world business scenarios, using a Mixture-of-Experts (MoE) architecture with 295B total parameters and 21B activated parameters. It natively supports a 256K context window and offers multiple thinking modes: no_think (ultra-fast response), think_low (quick thinking), and think_high (deep reasoning), balancing ultra-fast responses, complex reasoning, and invocation cost. Compared with the Preview version, Hy3—based on real business feedback from Tencent Yuanbao, WorkBuddy, ima, Marvis, and others—focuses on improving the Coding Agent, long-form understanding, multi-turn context continuity, search QA, and complex task execution, performing more stably in reducing hallucinations, improving task completion, and engineering usability. It is better suited to practical scenarios such as frontend tasks, cross-file code development, long-document analysis, office automation, and multi-step Agent workflows.

Using the Hunyuan Sheng 3D 3.1 model, it can generate higher-precision and higher-quality 3D models, supporting text-to-3D, image-to-3D, eight-view-to-3D, single-geometry generation (untextured model), sketch-to-3D, and intelligent topology-to-3D features.

Hunyuan-A13B-Instruct has 8 billion parameters and can match larger models by activating only 1.3 billion parameters, supporting "fast thinking/slow thinking" hybrid inference. It offers stable long text understanding. Verified by BFCL-v3 and τ-Bench, its Agent capabilities are leading in the field. Combined with GQA and multiple quantization formats, it enables efficient inference.

Hunyuan-MT-7B is a lightweight translation model with 7 billion parameters, designed to translate source text into target languages. The model supports translation among 33 languages as well as 5 Chinese minority languages. In the WMT25 International Machine Translation Competition, Hunyuan-MT-7B achieved first place in 30 out of 31 language categories it participated in, demonstrating its exceptional translation capabilities. For translation scenarios, Tencent Hunyuan proposed a complete training paradigm from pre-training to supervised fine-tuning, followed by translation reinforcement and ensemble reinforcement, enabling it to achieve industry-leading performance among models of similar scale. The model is computationally efficient, easy to deploy, and suitable for various application scenarios.