A German search engine tagged with tree planting and independence has replaced a prominent European AI company. Ecosia, headquartered in Berlin, recently announced a shift in its artificial intelligence strategy, ending its partnership with French Mistral and switching to various open-source solutions including Chinese open-source models. This move has caused quite a stir in the European tech community—because just a few months earlier, Ecosia regarded Mistral as a life-saving option to get rid of its dependence on U.S. giants.
Ecosia has always integrated environmental concepts and technical independence into the core of its brand. Its products have been used by public institutions such as the German Federal Ministry for the Environment and the UK National Health Service (NHS). In May this year, it replaced OpenAI with Mistral as its main AI supplier, aiming to build a more autonomous AI ecosystem using European local technology. Little did anyone expect that half a year later, this “European partnership” turned red.
Founder and CEO Christian Kroll explained the reasons for the breakup very clearly: The company was not satisfied with the overall performance of the Mistral model. It believed that its current model capabilities were about a year behind those of its main competitors, and it could not meet the growth demands of Ecosia. To make matters worse, technical issues occurred frequently during the partnership – Mistral’s servers overloaded multiple times, and service stability was compromised. Kroll even said bluntly that Ecosia was “too large” for Mistral, and its existing infrastructure was simply unable to handle its traffic.
More subtle is the mismatch at the value level. Ecosia questions whether Mistral truly aligns with its long-term strategy: the proportion of nuclear energy in France’s energy structure is relatively high, which somewhat conflicts with the company’s environmental narrative; it even doubts whether Mistral can represent genuine European technological sovereignty, as it also relies on international capital to survive. In Croel’s view, this approach cannot be considered technological independence.
Facing the public complaints from former customers, Mistral’s co-founder and chief scientist, Guillaume Lambour, did not remain silent. He publicly called on Ecosia to test Mistral’s latest model and promised to immediately open early access privileges; he emphasized that developing advanced AI requires time, computing power, and scientific research, and that Mistral is in a rapid growth phase, as well as capable of customizing models according to customer needs. This response shows reluctance, but also indirectly confirms the misalignment in the pace of both parties—one wants a stable, ready-made solution quickly, while the other is still in the process of growing.
Ecosia has pushed its chips towards Chinese open-source models, which represents a shift in the global AI market landscape. As China’s capabilities in open-source large models continue to improve, more and more overseas companies are incorporating them into their options. For Ecosia, performance, stability, and cost-effectiveness have surpassed the geographical factor as key considerations. Analysts believe this deal is symbolic for the European AI industry: Mistral, once seen as a leader in Europe, now has to cope with the pressure from US giants while also keeping up with the advances of Chinese open-source models. Ecosia’s vote-based system shows one thing—in the business market, companies will ultimately focus on technologies that offer the best overall performance, rather than simply paying for regional advantages.