Frontline Briefing: Tencent Hunyuan Hits 1.2 Billion Daily API Calls, Kimi Surpasses 37.6 Billion in Funding — China's AI Enters Capital-Technology Dual Acceleration

In mid-May 2026, a cluster of data points from China's AI industry captured global attention: Tencent's Hunyuan model reached 1.2 billion daily API calls, becoming the world's most active LLM; Moonshot AI (Kimi) surpassed 37.6 billion RMB in cumulative funding; DeepSeek V4.1 was confirmed for a June release with native multimodal capabilities; and Anthropic secured full access to SpaceX's Colossus 1 supercomputer.

What 1.2 Billion Daily Calls Really Means

Tencent Hunyuan's calls span dozens of products — WeChat Search, Tencent Docs, Tencent Meeting, QQ Browser — covering social, office productivity, search, and content generation. Unlike OpenAI's ChatGPT-centric distribution, Hunyuan's strength lies in scenario penetration. Users may not know it's Hunyuan, but they use its inference capabilities daily.

More importantly, these aren't "novelty" queries. In Q1 2026, Tencent reported AI-driven ad recommendations improved click-through rates by 18%, while AI code generation covered over 60% of internal R&D teams. Hunyuan has entered production-grade deployment.

Kimi's 37.6 Billion: Capital as Competitive Moat

Kimi's latest $2 billion round brought cumulative funding past 37.6 billion RMB, with a post-money valuation exceeding $20 billion. Achieved without Microsoft or Google backing — purely on model capability and developer ecosystem. Kimi's K2.6 recently beat Claude and GPT-5.5 in coding benchmarks, validating the capital with genuine technical competitiveness.

DeepSeek's Multimodal Sprint

DeepSeek V4.1 ships in June with image and audio understanding, just three months after V4. Meanwhile, Token pricing was slashed significantly, reshaping industry pricing. DeepSeek is building not just models but a complete "model + toolchain + protocol" ecosystem.

The Bigger Picture: China's Dual Acceleration

Capital acceleration (Kimi 37.6B + DeepSeek 50B RMB funding) plus usage acceleration (Hunyuan 1.2B/day + domestic weekly 7.94T tokens) = China's AI sector has entered a historically unprecedented dual-acceleration mode. The question is no longer "can China catch up" — it's "when does a genuine dual-center global AI landscape emerge?"