Ever since Microsoft transitioned its Copilot Cowork offering to a metered pricing architecture, where you pay for the number of tokens consumed, it has been eyeing open-source models from China with particular interest, especially given the materially lower inference costs that such models entail.
Now, a new report suggests that Microsoft might add Moonshot's Kimi K3 model to Copilot, risking the ire of the Trump administration in the process.
Microsoft is working to add Kimi K3 model to its Azure cloud services, while its engineers continue to evaluate the model for a deployment within Copilot
According to The Information, Microsoft is already working to offer access to Moonshot's Kimi K3 model via its Azure cloud services. What's more, the Redmond giant is also evaluating the model's efficacy for deployment within Copilot.
This follows an Axios report back in June that suggested Microsoft might opt for DeeSeek's V4 model, or another similar one, albeit hosted on its own infrastructure, to power Copliot Cowork, which combines Copilot's enterprise-related functions with advanced AI models from OpenAI and Anthropic to make agentic AI-powered workflows a relative breeze.
As we explained in a recent post, Moonshot has just unveiled its brand-new, open-source Kimi K3 model, which spans 2.8 trillion parameters, and is designed specifically for frontier-scale intelligence.
The model is multi-modal, has a context window of around 1 million tokens, and offers competitive and faster performance than many of its compeers.
Even so, Kimi K3's inference costs are not exactly cheap for an open-source model, entailing an average cost of $0.94 per Intelligence Index task versus $0.55 for GPT-5.6 Terra and $1.04 for GPT-5.6 Sol, and that too with maximum reasoning enabled. What's more, Kimi K3 also has a propensity to reason more, which consumes additional tokens.
Yet, as we explained in a dedicated post recently, Kimi K3 is quite efficient to run on data center-scale hardware, courtesy of its dramatically reduced KV cache and built-in optimizations for per-token memory use - achieved by distributing its 896 experts across a large number of GPUs.
In fact, Microsoft calculates that it could save as much as $600 million in inference costs by moving Copilot away from OpenAI and Anthropic models and towards Moonshot's Kimi K3.
Of course, such a move is unlikely to sit well with the Trump administration, which is trying to discourage the use of open-source models from China. But, with the second US AI security chief now having resigned in the past three months or so, the administration's overarching AI strategy currently lies in a disarray.
About the author: Writing is my one incontrovertible passion. Over the past six years, he has authored over 2,200 distinct articles on financial and tech-related topics, spanning nearly 1 million words. And he has been a member of Wcctech mobile team since 2025. As an alumnus of the University of Toronto, Rotman Commerce Program, I bring nuance, in-depth knowledge, and a unique perspective to every topic that I cover. When I'm not writing, I'm traveling the world, exploring hidden confectionaries and restaurants as an aspiring food connoisseur.
Follow Wccftech on Google to get more of our news coverage in your feeds.
