Nvidia Expands Support for Chinese Open-Source AI Models While Warning of Export Control Risks
Summary
- • Nvidia offers day-zero hardware support for DeepSeek, Qwen, and other Chinese open-source AI models
- • SEC filing warns US restrictions on Chinese AI models could materially harm Nvidia's business
- • RTX GPUs support Qwen3.8-27B; DGX Spark clusters optimized for DeepSeek V4 Flash and GLM 5.2
- • Nvidia says supporting Chinese models keeps global developers on the American tech stack
Details
Nvidia launches Local AI Initiative for Chinese open-source models
Nvidia's program provides day-zero hardware and software support for leading Chinese open-source AI models on RTX and DGX systems, part of a broader local AI strategy.
Day-zero RTX GPU support enabled for Alibaba Qwen3.8-27B
RTX GPU systems receive day-zero software support for Alibaba's Qwen3.8-27B as part of the local AI initiative, enabling on-device inference out of the box.
DGX Spark clusters optimized for DeepSeek V4 Flash and Z.ai GLM 5.2
Software updates enable clustering of DGX Spark systems to run DeepSeek V4 Flash and Z.ai's GLM 5.2, extending Nvidia's support to leading Chinese open-weight models.
SEC filing warns Chinese AI restrictions could materially harm Nvidia
Nvidia's Q2 SEC filing warned that potential US government restrictions on third-party apps built on Chinese open-source models—specifically naming DeepSeek, Qwen, and Kimi—could have a material adverse business impact.
Nvidia argues supporting Chinese models preserves American tech stack dominance
An unnamed Nvidia employee told CNBC that supporting global models, including Chinese ones, ensures developers continue building on American infrastructure rather than switching to Chinese alternatives.
Chinese chipmakers Huawei and Alibaba also optimizing for the same open-source models
Huawei's Ascend chips and Alibaba hardware also announced support optimizations for leading Chinese open-source models, intensifying competition for AI developer workloads.
Nvidia's dual move—actively supporting Chinese open-source AI models commercially while flagging regulatory risk in its SEC filing—highlights the deep entanglement of US hardware and Chinese AI software ecosystems.
What This Means
Nvidia is actively competing for Chinese open-source AI adoption by providing day-zero hardware support for models like DeepSeek and Qwen, even as US lawmakers grow increasingly concerned about Chinese AI proliferation. The company's simultaneous SEC warning signals real regulatory exposure—restrictions on Chinese open-source model support could materially impact Nvidia's business. This reveals the deep entanglement of US hardware and Chinese AI software ecosystems, making a clean policy separation extremely difficult without significant collateral damage to American chipmakers. Nvidia's public "American tech stack" justification for supporting Chinese models will face increasing scrutiny as US-China AI rivalry intensifies.