# Tencent AI

> Source: https://aiwiki.ai/wiki/tencent_ai
> Updated: 2026-06-21
> Categories: AI Companies, Chinese AI, Large Language Models
> From AI Wiki (https://aiwiki.ai), a free encyclopedia of artificial intelligence. Quote with attribution.

**Tencent AI** is the artificial intelligence research, products, and services developed by Tencent Holdings Ltd., one of the world's largest technology companies, built around the **Hunyuan** family of foundation models and the **Yuanbao** consumer assistant. Headquartered in Shenzhen, China, Tencent integrates these models into products used by over a billion people, including [WeChat](/wiki/wechat), QQ, [Tencent Cloud](/wiki/cloud_computing), and Yuanbao. The Hunyuan lineup spans [large language models](/wiki/large_language_model), image generators, [video generation](/wiki/ai_video_generation) systems, and 3D asset creation tools, and by early 2026 the AI-powered search built on Hunyuan handled more than 90% of all question-based queries inside WeChat.[2][14] Tencent pledged to more than double its dedicated AI spending in 2026 to over CNY 36 billion (about $5.2 billion), part of a 2025 capital expenditure of roughly CNY 79 billion ($11.49 billion).[17][18]

Tencent was [co-founded](/wiki/artificial_intelligence) in 1998 and reported 2025 revenue of CNY 751.8 billion (about $108.8 billion), up 14% year over year, with net profit up 16% to CNY 224.8 billion.[15]

## Company Background

Tencent was co-founded on November 11, 1998, by Ma Huateng (known globally as Pony Ma), Zhang Zhidong, Xu Chenye, Chen Yidan, and Zeng Liqing in Shenzhen, Guangdong Province, China.[21] The company name "Tencent" derives from its Chinese name Tengxun (腾讯), which incorporates part of Ma Huateng's Chinese name and loosely translates to "galloping fast information."[21] Ma Huateng graduated from Shenzhen University in 1993 with a degree in computer science and continues to serve as Chairman and CEO.[22]

Tencent's first product was OICQ, an instant messaging client inspired by ICQ, which launched in February 1999. The product was later renamed QQ following a trademark dispute with AOL.[21] From this foundation, Tencent expanded into gaming, social media, fintech, cloud computing, and digital entertainment, growing into a conglomerate with a market capitalization exceeding $500 billion as of early 2026 and approximately 105,000 employees.[21]

## AI Research History

Tencent's investment in artificial intelligence predates the generative AI boom by more than a decade. The company established multiple dedicated AI research organizations over the years.

### Tencent Research Institute (2007)

In 2007, Tencent invested over RMB 100 million to establish the Tencent Research Institute, with campuses in Beijing, Shanghai, and Shenzhen.[21] This was one of the company's earliest formal commitments to long-term technology research.

### Tencent YouTu Lab (2012)

Tencent YouTu Lab was founded in 2012 and became one of Tencent's three core AI laboratories. Based in Shanghai and Hefei, YouTu Lab specializes in [computer vision](/wiki/computer_vision), including [optical character recognition](/wiki/optical_character_recognition), [image recognition](/wiki/image_recognition), and image generation. The lab developed AI capabilities for internal products such as QQ and Qzone, as well as for external partners including WeBank, SF Express, and China Unicom.

### Tencent AI Lab (2016)

Tencent AI Lab was formally established in April 2016 after preparations that began in 2015. The lab recruited more than 50 AI scientists and 200 AI application engineers. In March 2017, Tong Zhang, an expert in [machine learning](/wiki/machine_learning) and former professor at Rutgers University, became the lab's director, and Dong Yu, a former Microsoft researcher specializing in [speech recognition](/wiki/speech_recognition), was appointed vice director.

Tencent AI Lab focuses on four core research areas: computer vision, speech technology, [natural language processing](/wiki/natural_language_processing), and machine learning. In 2017, the lab opened its first U.S. research center in Seattle to advance work on speech recognition and NLP.

In March 2026, Tencent announced a major organizational restructuring, folding the AI Lab into the Hunyuan team to consolidate its AI efforts under a single, more focused unit.[16]

## Hunyuan Foundation Model

### When was Hunyuan first launched?

Tencent officially unveiled the Hunyuan large language model on September 7, 2023, at the company's Global Digital Ecosystem Summit.[1][2] The initial model featured over 100 billion parameters and was pre-trained on 2 trillion tokens.[2] Dowson Tong, Tencent's Senior Executive Vice President, described the model as having "strong capabilities in Chinese language comprehension, content creation, logical reasoning, and task execution."[2]

At launch, Hunyuan was made available to domestic enterprise users through APIs on [Tencent Cloud](/wiki/cloud_computing), supporting a Model-as-a-Service (MaaS) offering.[2] The consumer-facing chatbot remained in internal testing on an invite-only basis.[1] Tencent also began embedding Hunyuan into several of its own products, including Tencent Cloud, Tencent Games, and WeChat.[2]

### Hunyuan-Large / Hunyuan-MoE-A52B (November 2024)

On November 4, 2024, Tencent open-sourced Hunyuan-Large, also known as Hunyuan-MoE-A52B.[3][4] This was a [Transformer](/wiki/transformer)-based [mixture of experts](/wiki/mixture_of_experts) model with 389 billion total parameters and 52 billion activated parameters per forward pass, described at release as the largest open-source Transformer-based MoE model to date.[3][4] The model supports a context window of up to 256,000 tokens.[3]

Hunyuan-Large was pre-trained on 7 trillion tokens, including 1.5 trillion tokens of synthetic data designed to improve performance in mathematics, coding, and multilingual tasks.[3] Key technical innovations included a mixed expert routing strategy, key-value cache compression, and expert-specific learning rate scheduling.[3]

On benchmarks, Hunyuan-Large outperformed [LLaMA](/wiki/llama) 3.1-70B across multiple evaluations and demonstrated comparable performance to the much larger [LLaMA 3](/wiki/llama_3).1-405B model.[3] On [MMLU](/wiki/mmlu), Hunyuan-Large scored 88.4, compared to LLaMA 3.1-405B's 85.2 and LLaMA 3.1-70B's 79.3.[3][4] The model weights were released on Hugging Face in three variants: Hunyuan-A52B-Pretrain, Hunyuan-A52B-Instruct, and Hunyuan-A52B-Instruct-FP8.[4]

### Hunyuan TurboS (February 2025)

On February 27, 2025, Tencent released Hunyuan-TurboS, the first ultra-large Hybrid-Transformer-[Mamba](/wiki/mamba) MoE model deployed in a production environment.[19] The model was scaled to 560 billion total parameters with 56 billion activated parameters, comprising 128 layers in an architecture that alternates between [Attention](/wiki/attention), Mamba2, and FFN blocks.[19]

The Mamba2 layers use a state-space model (SSM) architecture with linear sequence-length complexity, addressing the quadratic scaling challenges of pure Transformer models in long-context scenarios.[19] Each FFN layer uses an MoE structure with 1 shared expert and 32 specialized experts, activating 1 shared plus 2 specialized experts per forward pass.[19] The model was pre-trained on 16 trillion high-quality tokens and supports context lengths up to 256,000 tokens.[19] TurboS also introduced an adaptive long-short chain-of-thought mechanism that switches between rapid responses for simple queries and deeper "thinking" for complex ones.[19]

Hunyuan-TurboS reached an overall top-8 ranking on the LMSYS [Chatbot Arena](/wiki/lmsys_chatbot_arena) with a score of 1356, comparable to Gemini-2.0-Flash-001 (1352), and averaged 77.9% across 23 automated benchmarks.[19]

### Hunyuan T1 Reasoning Model (March 2025)

In mid-February 2025, Tencent launched the Hunyuan T1-Preview reasoning model on the Yuanbao app, followed by the official T1 release on March 21, 2025.[7][8] Built on the TurboS base, T1 activates 52 billion parameters through dynamic expert routing, with specialist modules handling specific reasoning domains like mathematical logic and contextual analysis.[8]

T1 achieved strong results on reasoning benchmarks: 87.2 on MMLU-PRO (second only to [OpenAI](/wiki/openai) o1), 69.3 on [GPQA](/wiki/gpqa)-Diamond for scientific reasoning, 64.9 on [LiveCodeBench](/wiki/livecodebench) for coding, and 96.2 on MATH-500 for mathematical reasoning.[8] The model generates at 60 to 80 tokens per second and charges 1 yuan (approximately $0.14) per million input tokens and 4 yuan per million output tokens.[7]

### Hunyuan 2.0 (December 2025)

Tencent released Hunyuan 2.0 on December 5, 2025.[9][10] This MoE model features 406 billion total parameters with 32 billion active parameters and supports a 256K context window.[9] The model ships in two variants: Think (optimized for complex reasoning) and Instruct (optimized for instruction following).[9]

Key improvements include higher-quality pre-training data and a dual-stage [reinforcement learning](/wiki/reinforcement_learning) approach combining RLVR (Reinforcement Learning with Verifiable Rewards) and [RLHF](/wiki/rlhf) (Reinforcement Learning from Human Feedback).[9] On the IMO-AnswerBench mathematics benchmark, Hunyuan 2.0 Think scored 73.4. The Tau2-Bench agent benchmark score jumped from 17.1 (in the previous generation) to 72.4.[9][10] At launch, Hunyuan 2.0 was already integrated into Tencent's AI-native products, including Yuanbao and ima, and offered through Tencent Cloud APIs.[10]

### Hunyuan 3.0 (Planned April 2026)

As of March 2026, Tencent has announced plans to launch Hunyuan 3.0 in April 2026, alongside an advanced AI agent for WeChat.[15] An earlier open-sourced Hunyuan 3.0 preview model used a [mixture of experts](/wiki/mixture_of_experts) design with 295 billion total parameters and only 21 billion active parameters, a design philosophy that prioritized inference efficiency and cost optimization over trillion-parameter scale.[15] The planned WeChat agent is designed to let users invoke mini-programs and complete end-to-end services such as ride-hailing and food ordering directly inside the app.[15]

## Hunyuan Model Family Overview

| Model | Release Date | Type | Total Parameters | Active Parameters | Key Features |
|---|---|---|---|---|---|
| Hunyuan (v1) | September 2023 | LLM | 100B+ | Dense | Chinese/English language, 2T training tokens |
| Hunyuan-Large (MoE-A52B) | November 2024 | LLM (MoE) | 389B | 52B | 7T training tokens, 256K context, open-source, MMLU 88.4 |
| HunyuanDiT | May 2024 | Image Generation | N/A | N/A | Bilingual diffusion transformer, multi-turn generation |
| HunyuanVideo | December 2024 | Video Generation | 13B | Dense | Text-to-video, open-source, 13B parameters |
| Hunyuan TurboS | February 2025 | LLM (Hybrid MoE) | 560B | 56B | Mamba-Transformer hybrid, 16T training tokens, Arena 1356 |
| Hunyuan T1 | March 2025 | Reasoning LLM | N/A | 52B | Deep reasoning, 60-80 tokens/sec generation |
| HunyuanImage 3.0 | September 2025 | Image Generation | 80B | 13B | Autoregressive MoE, 64 experts, bilingual |
| HunyuanVideo 1.5 | November 2025 | Video Generation | 8.3B | Dense | Lightweight, runs on consumer GPUs |
| Hunyuan 2.0 | December 2025 | LLM (MoE) | 406B | 32B | Think/Instruct variants, 256K context |
| Hunyuan3D 2.1 | 2025 | 3D Generation | N/A | N/A | Text/image-to-3D with PBR materials |
| Hunyuan 3.0 (preview) | April 2026 (planned) | LLM (MoE) | 295B | 21B | In-context learning, agent focus, efficiency-first |

## HunyuanDiT: Image Generation

HunyuanDiT (Hunyuan Diffusion Transformer) is Tencent's text-to-image model, open-sourced in May 2024.[12] The model uses a [diffusion](/wiki/diffusion_models) architecture parameterized with a transformer and combines a pre-trained bilingual [CLIP](/wiki/clip) encoder with a multilingual [T5](/wiki/t5) encoder for text processing.[12]

A distinguishing feature of HunyuanDiT is its strong bilingual support for both Chinese and English, with particular strength in rendering Chinese text within generated images.[12] The model also supports multi-turn text-to-image generation, allowing users to refine outputs through iterative dialogue.[12] An integrated multimodal large language model interprets multi-round conversations and generates updated prompts for image generation.[12]

Subsequent versions included HunyuanDiT v1.2 (July 2024), which introduced a low-VRAM variant requiring only 6 GB of GPU memory, and integration with the Diffusers library in June 2024.

### HunyuanImage 3.0 (September 2025)

Tencent open-sourced HunyuanImage 3.0 on September 28, 2025.[20] Unlike HunyuanDiT's diffusion-based approach, HunyuanImage 3.0 uses a unified autoregressive framework similar to the architecture powering large language models.[20] The model features 80 billion total parameters with an MoE structure containing 64 experts, activating 13 billion parameters during inference.[20]

HunyuanImage 3.0 supports precise Chinese and English text rendering, handles prompts exceeding 1,000 characters, and produces high-resolution outputs.[20] An Instruct variant was released on January 26, 2026.

## HunyuanVideo: Video Generation

[HunyuanVideo](/wiki/hunyuan_video) is Tencent's open-source text-to-video generation model. The inference code and model weights were first released on December 3, 2024.[5][6] With over 13 billion parameters, HunyuanVideo was the largest open-source video generation model at its time of release.[5]

The model uses a "Dual-stream to Single-stream" hybrid transformer design with a Full Attention mechanism for unified image and video generation.[5] In professional human evaluations involving over 1,500 prompts assessed by 60 evaluators, HunyuanVideo outperformed [Runway](/wiki/runway_ml) Gen-3, Luma 1.6, and several leading Chinese video generation models.[5] It achieved a 64.5% motion quality score compared to Gen-3's 48.3%.[5]

HunyuanVideo was integrated into the Diffusers library on December 17, 2024, and FP8 model weights were released on December 18, 2024, to reduce GPU memory requirements.

### HunyuanVideo 1.5 (November 2025)

HunyuanVideo 1.5 was released in November 2025 as a more efficient successor.[11] The model reduced the parameter count from 13 billion to 8.3 billion while maintaining top-tier video quality.[11] A key technical innovation is Selective and Sliding Tile Attention (SSTA), which achieves a 1.87x speedup for 10-second 720p video generation compared to [FlashAttention](/wiki/flash_attention)-3.[11]

The 1.5 version is designed to run on consumer-grade GPUs.[11] A step-distilled 480p image-to-video model released on December 5, 2025, generates videos in 8 to 12 steps, reducing end-to-end generation time by 75% on an RTX 4090.[11] Training code and [LoRA](/wiki/lora) tuning scripts were also released for community fine-tuning.[11]

## Hunyuan3D: 3D Asset Generation

The Hunyuan3D series enables text-to-3D and image-to-3D generation. Since the first open-source 3D models were released in November 2024, the series has surpassed 3 million community downloads on [Hugging Face](/wiki/hugging_face).

**Hunyuan3D 2.0** introduced a two-component system: Hunyuan3D-DiT for large-scale shape generation and Hunyuan3D-Paint for texture synthesis. **Hunyuan3D 2.1** became the first production-ready 3D asset generation model, with full model weights and training code released for community fine-tuning. It supports PBR (Physically Based Rendering) materials for production workflows.

**HunyuanWorld 1.0**, released on July 26, 2025, is described as the industry's first open-source 3D world generation model. It generates immersive, explorable 3D environments from text or image inputs and is compatible with standard CG pipelines.

## Yuanbao: Consumer AI Assistant

Tencent launched the Yuanbao AI assistant app on May 30, 2024, as its consumer-facing [chatbot](/wiki/chatbot) powered by Hunyuan.[13] Yuanbao provides AI-powered search (integrating WeChat Search and Sogou Search), document analysis (supporting PDF, Word, Excel, and other formats), content creation, and image editing features.[13]

Yuanbao was subsequently integrated directly into WeChat, allowing users to add it as a "friend" within the app and interact with it without downloading a separate application.[14] This integration provides over a billion WeChat users with direct access to Tencent's AI capabilities.[14] By early 2026, the AI-powered search function within WeChat handled over 90% of all question-based queries on the platform.[14] Driven by this ecosystem placement, Yuanbao became the fastest-growing AI assistant in China, reaching 41.64 million monthly active users by the second quarter of 2025.[14]

Tencent also introduced Yuanbao Groups in January 2026 for AI-powered social interactions, and the Yuanbao chatbot supports summarizing WeChat official account articles, responding to images and files, and generating structured reports.

## Product and Platform Integration

Tencent has embedded Hunyuan across a broad range of products and services.

### WeChat and QQ

WeChat users gained access to AI features through the Yuanbao integration, initially appearing as a "Red Packet Cover Assistant" before expanding to broader assistant capabilities.[14] QQ's browser was upgraded to an "AI Browser" with QBot, an assistant built on both Hunyuan and [DeepSeek](/wiki/deepseek) models, offering AI-powered search, office productivity, learning, and writing functions.

Ima.copilot, an AI-native application from the QQ Browser team, offers integration with WeChat Official Accounts and instant messaging. Sogou Input Method, also owned by Tencent, has integrated Hunyuan for enhanced text prediction and composition.

### Tencent Meeting and Tencent Docs

Tencent Meeting now incorporates intelligent recording, an AI assistant, and multilingual translation, serving over 15 million monthly users.[23] Tencent Docs similarly uses Hunyuan for document generation and editing assistance.[23]

### AI Coding Assistant

Tencent Cloud's AI Coding Assistant, powered by Hunyuan, is used by more than 50% of Tencent's internal programmers and reportedly enhances productivity by 40%.[23]

## Tencent Cloud AI Services

Tencent Cloud offers AI capabilities to enterprise customers through several products and platforms.

### Model-as-a-Service (MaaS)

Tencent Cloud provides access to Hunyuan foundation models via APIs, enabling businesses to build and fine-tune industry-specific models.[23] The MaaS offering includes more than 50 solutions across 20 major industries.[23] Tencent also released Hunyuan Turbo as a cost-optimized model service, which doubled training efficiency and reduced inference costs by 50% compared to earlier offerings.[23]

### Agent Development Platform 3.0

Launched as part of Tencent Cloud's enterprise AI suite, [Agent](/wiki/agent) Development Platform 3.0 (ADP) enables organizations to build and integrate autonomous [AI agents](/wiki/ai_agents) for customer service, marketing, inventory management, and research workflows.[23]

### Tairos: Embodied Intelligence

Tencent unveiled Tairos, an embodied intelligence platform designed to serve as the "AI brain" for humanoid robots and other physical AI systems.[23] Tairos provides perception, motion planning, and human-machine interaction capabilities for robotics developers.[23]

### AI Infrastructure

Tencent Cloud introduced a full suite of computing, storage, and networking solutions branded as AI Infra, optimized for large-scale model training and inference workloads.[23]

## Competition in China's AI Market

Tencent competes in one of the world's most crowded AI markets. As of April 2025, more than 3,700 registered generative AI tools operated in China. Tencent's primary competitors include:

| Company | Primary LLM | Consumer Chatbot | Key Differentiator |
|---|---|---|---|
| [Baidu](/wiki/baidu_ai) | [ERNIE](/wiki/ernie) | Ernie Bot | Early mover in search-integrated AI |
| [Alibaba](/wiki/alibaba_cloud) | [Qwen](/wiki/qwen) | Tongyi Qianwen | Aggressive open-source strategy, e-commerce integration |
| [ByteDance](/wiki/bytedance_ai) | Doubao / Seed | Doubao | Short-video ecosystem (Douyin/TikTok) |
| [DeepSeek](/wiki/deepseek) | DeepSeek R1/V3 | DeepSeek Chat | Cost-efficient reasoning models |
| Tencent | Hunyuan | Yuanbao | WeChat ecosystem, social integration |

### How big is Yuanbao compared with rivals?

Despite being the fastest-growing assistant in China, Yuanbao still trailed the market leaders in raw scale as of 2025: DeepSeek reported roughly 194 million monthly active users and Doubao about 116 million, compared with Yuanbao's 41.64 million.[14] Tencent's advantage is distribution rather than standalone app downloads: Yuanbao can reach WeChat's user base and its trove of official-account content, reducing reliance on external search engines.[14]

### The Price War

China's AI market has been characterized by aggressive price competition. In May 2024, ByteDance released models costing roughly 99% less than [OpenAI](/wiki/openai)'s comparable offerings. Alibaba followed by cutting Qwen model prices by 97%. In December 2024, Alibaba reduced Qwen-VL pricing by 85%, and in April 2025, Baidu cut Ernie 4.5 Turbo pricing by 80%.

### User Acquisition Spending

During the 2026 Chinese New Year period, major tech companies spent heavily to acquire AI chatbot users. Baidu allocated CNY 500 million (approximately $72 million) to promote Ernie Bot. Tencent doubled that figure, spending CNY 1 billion (approximately $145 million) on Yuanbao promotion. Alibaba outspent both with CNY 3 billion (approximately $434 million) for Qwen adoption.

### Strategic Positioning

Tencent's competitive advantage in AI stems from its dominant social ecosystem. WeChat serves over 1 billion monthly active users, giving Tencent unmatched distribution for AI features.[14] Bloomberg Intelligence has noted that Tencent and Alibaba are positioned to lead China's AI software sector, outpacing Baidu as [DeepSeek](/wiki/deepseek) enables broader tech development. Tencent has focused on embedding AI into existing products rather than competing solely on model benchmarks, using AI to enhance advertising targeting, content recommendations, and enterprise productivity.

## Investment and Future Direction

### How much is Tencent spending on AI?

Tencent's total capital expenditure reached approximately 79 billion yuan ($11.49 billion) in 2025, up from 77 billion yuan in 2024.[18] However, spending fell below internal targets due to constraints in acquiring advanced AI chips caused by U.S. export controls.[18]

For 2026, Tencent plans to more than double its AI investment, with spending expected to exceed CNY 36 billion ($5.2 billion) specifically on AI.[17][18] Goldman Sachs projects total Tencent capital expenditure will reach RMB 100 billion in 2026 and RMB 110 billion in 2027.[17] In the fourth quarter of 2025 alone, investment in Yuanbao and Hunyuan large models reached 1.6 billion RMB, with annual spending totaling 1.8 billion RMB.[17] Chief strategy officer James Mitchell pledged "a substantial increase" in 2026 capital expenditure, especially in the second half, as more China-designed AI chips become available.[18]

Tencent's strategy for 2026 centers on three priorities: upgrading Hunyuan foundation models (with Hunyuan 3.0 planned for April), deepening AI integration across WeChat and other consumer products through AI agents, and expanding Tencent Cloud's AI infrastructure globally with localized, sovereign computing resources.[15]

## Open-Source Contributions

### Is Hunyuan open source?

Tencent has adopted a substantial open-source strategy for many of its Hunyuan models. The following models have been released with open weights on platforms like GitHub and Hugging Face:

- **Hunyuan-Large** (November 2024): The largest open-source MoE model at its time of release
- **HunyuanDiT** (May 2024): Bilingual text-to-image diffusion transformer
- **HunyuanVideo** (December 2024): 13B-parameter text-to-video model
- **HunyuanVideo 1.5** (November 2025): Lightweight 8.3B video generation model with training code
- **Hunyuan3D series** (November 2024 onward): Text-to-3D and image-to-3D models with PBR support
- **HunyuanImage 3.0** (September 2025): 80B-parameter autoregressive image generation model
- **HunyuanWorld 1.0** (July 2025): First open-source 3D world generation model
- **Hunyuan small LLMs** (July 2025): Efficient models in 0.5B, 1.8B, 4B, and 7B parameter sizes

These releases are generally made available under the Tencent Hunyuan Community License Agreement, which permits research, educational, and commercial use with attribution requirements.

## References

1. "Tencent launches Hunyuan large language model." TechNode, September 8, 2023.
2. "Tencent Unveils Hunyuan, its Proprietary Large Foundation Model on Tencent Cloud." Tencent Official Press Release, September 2023.
3. "Hunyuan-Large: An Open-Source MoE Model with 52 Billion Activated Parameters by Tencent." arXiv:2411.02265, November 4, 2024.
4. "Tencent Releases Hunyuan-Large (Hunyuan-MoE-A52B) Model." MarkTechPost, November 5, 2024.
5. "HunyuanVideo: A Systematic Framework For Large Video Generative Models." arXiv:2412.03603, December 3, 2024.
6. "Tencent Open Sources HunyuanVideo." [ComfyUI](/wiki/comfyui) Wiki, December 1, 2024.
7. "Tencent's Hunyuan T1 AI reasoning model rivals DeepSeek in performance and price." South China Morning Post, March 2025.
8. "Tencent Releases Hunyuan-T1: First Ultra-Large Mamba-Powered Language Model." Learn Prompting, March 2025.
9. "Tencent Releases Hunyuan 2.0 With 406B Parameters." Dataconomy, December 8, 2025.
10. "Tencent releases Hunyuan 2.0, its next-generation AI model." TechNode, December 8, 2025.
11. "HunyuanVideo 1.5 Technical Report." arXiv:2511.18870, November 2025.
12. "Hunyuan-DiT: A Powerful Multi-Resolution Diffusion Transformer with Fine-Grained Chinese Understanding." arXiv:2405.08748, May 2024.
13. "Tencent launches Yuanbao AI assistant app." South China Morning Post, May 30, 2024.
14. "Tencent adds AI chatbot friend to WeChat." South China Morning Post, 2025.
15. "Tencent to Launch Hunyuan 3.0 in April, Build WeChat AI Agent." Caixin Global, March 18, 2026.
16. "Tencent Folds AI Lab Into Hunyuan Team in Major AI Overhaul." Caixin Global, March 21, 2026.
17. "Tencent to double AI spending on 13% revenue growth as chip curbs reshape capex." Digitimes, March 2026.
18. "Tencent pledges higher AI investment in 2026 after chip curbs hit capex plans." Reuters, March 2026.
19. "Hunyuan-TurboS: Advancing Large Language Models through Mamba-Transformer Synergy and Adaptive Chain-of-Thought." arXiv:2505.15431, 2025.
20. "Hunyuan Image 3.0 Complete Guide: Tencent's 80B Parameter AI Model." WaveSpeed AI Blog, 2026.
21. "Tencent." Wikipedia, accessed March 2026.
22. "Ma Huateng." Britannica Money, accessed March 2026.
23. "Tencent Announces Global Rollout of Scenario-Based AI Capabilities." Tencent Official Press Release, 2025.