Moonshot AI has made its latest model Kimi K3 available for public download, releasing the model's weights alongside key infrastructure tools designed to improve efficiency and stability2. The open-source release and the model's support for non-Nvidia ecosystems have triggered heated debates across Silicon Valley over potential bans on Chinese open-source models.
Moonshot's 2.8-trillion-parameter model prompted broad reaction across global technology markets when it was first released, intensifying debate over the competitive positioning of Western AI labs ctx. Chip stocks entered a bear market on Friday, driven in part by questions about the model's scale ctx.
The policy dimension is sharpening in parallel. U.S. Treasury Secretary Scott Bessent said the Trump administration will investigate whether Chinese AI models have been distilled from American models, threatening sanctions against Chinese AI companies if intellectual property theft is established ctx. The Financial Times has raised the question of what risks accompany the use of Chinese open AI models like Kimi K31.
Moonshot AI is also reportedly planning a final pre-IPO funding round targeting a pre-money valuation of up to $50 billion, with talks expected to begin in August ctx.
ANALYSIS By releasing not only model weights but also supporting infrastructure, Moonshot AI is lowering the barrier for developers worldwide to adopt Kimi K3 — a move that directly tests the boundaries of any proposed restrictions on Chinese open-source AI. The combination of an open-weight release, explicit non-Nvidia hardware support, and a looming U.S. sanctions investigation creates a concrete collision between open-source distribution norms and national-security policy frameworks.