Sparxity Completes Series A, Bringing Total Funding to €80 Million

eu-startups.com reported on November 10, 2026, that Dresden-based Sparxity had completed its Series A, bringing its total funding to €80 million ($90 million). The amount raised in this round alone was not disclosed. The company reportedly changed its name from SpiNNcloud. The round was led by ICM HPQC Fund, with GlobalFoundries, Lockheed Martin Ventures, GS Futures and others participating.

eu-startups.com reported that Sparxity envisions heterogeneous data centers equipped with dedicated silicon for ultra-sparse processing. If realized, the concept could be relevant to companies considering how to allocate computing resources based on processing characteristics, rather than simply increasing general-purpose computing capacity. The report did not provide performance figures or operational track records.

The €80 million figure is the company’s cumulative funding, not the size of the Series A alone; it includes previous funding. The announcement therefore does not reveal the amount raised in this round or the company’s valuation. The business changes confirmed in the report are the completion of the funding round and the name change from SpiNNcloud to Sparxity. The report does not say whether the name change involves changes to its products or business structure. The company is described as focused on sparse-native computing. According to the report, it aims to build heterogeneous data centers incorporating dedicated silicon for ultra-sparse processing. This suggests that the focus is not only on the total amount of computing resources, but also on designing systems that use different resources according to processing characteristics. However, the report does not provide chip specifications, supported workloads, or comparisons of performance or power efficiency with existing computing platforms. The technology’s effectiveness cannot be judged from the funding amount alone. In related coverage, Google Cloud AI Research Releases Harness²; Google DeepMind Research Raises Repair Rate From 24.4% to 48.3% examined research into how the setup of execution environments affects software results. That concerns software-side techniques and differs from Sparxity’s proposed computing hardware. Separately, Google DeepMind Unveils ‘Genie 4,’ an AI for Generating Virtual Worlds, With Audio and Conversation Support covered the generation of virtual environments intended for uses such as robot training. Developing models and environments and building the computing infrastructure that runs them are distinct layers; the information available does not establish which workloads might use Sparxity’s proposed system. The listed investors include GlobalFoundries and Lockheed Martin Ventures, as well as investment funds and venture investors. The report names the participants but does not disclose their individual contributions or their technical or business involvement, so it does not provide grounds to infer partnerships or supply relationships. Points to watch include the amount raised in the Series A alone, the specifications of the silicon under development, measured performance and power consumption, and its operating status in data centers. The report provides no figures or deployment record with which to assess these.