Sparxity Closes Series A, Bringing Total Funding to $90 Million for Sparse AI Computing

Sparxity, the Dresden-based computing company formerly known as SpiNNcloud, has closed a Series A financing that brings cumulative funding secured by its team to $90 million. The investment will support specialized chips designed for fine-grained dynamic sparsity, an approach that concentrates AI computation on the parts of a model needed for each task.

The round was led by ICM HPQC Fund, with investors including Lockheed Martin Ventures, GlobalFoundries, GS Futures, Cloudberry Ventures, TGFS, and the EIC Fund. Angel investors include Richard Socher and Nancy Xu, alongside continued backing from Atlantic and Silicon Catalyst. Sparxity did not disclose the Series A amount separately from the cumulative funding figure.

Its ambition is to make increasingly sparse AI models practical to run at scale. The next-generation hardware targets up to 44 times faster inference in ultra-sparse models, building on an existing commercial business in sparse computing systems.

Why Sparse Models Need Different Hardware

In a dense neural network, substantial computation is applied across a large set of parameters. Sparse architectures selectively activate parts of a model, potentially increasing its overall capacity without requiring every parameter to participate in every calculation.

The opportunity becomes harder to realize when that selection is highly granular and changes with the input. Hardware must route work to the right resources and move the relevant data efficiently. Skipping arithmetic helps only if the cost of finding, scheduling, and communicating the remaining work does not erase the gain.

Sparxity’s technology overview frames its architecture around that problem, pointing to finer-grained selection across experts, memory, and attention. The company is developing chips and software together so model builders can use sparse structures that are awkward to execute efficiently on hardware designed primarily around dense computation.

“The next era of AI will be defined not just by how much computing power we can deliver, but by how intelligently and efficiently we use it,” said Hector Gonzalez, co-founder and CEO of Sparxity. He said the financing will accelerate dedicated silicon for ultra-sparse workloads within heterogeneous data centers.

From Millions of Experts to Selective Computation

One research direction behind the strategy is PEER, or parameter efficient expert retrieval. In Mixture of A Million Experts, researcher Xu Owen He described a layer that retrieves a small selection from a pool of more than a million tiny experts. Experiments on language-modeling tasks examined how that approach changes the relationship between model performance and computational cost.

The principle is to expand what a model can draw on without processing the entire pool for each input. Translating that model-level efficiency into lower latency and energy use requires a system that can handle the resulting fine-grained work.

He, a former Google DeepMind researcher and Sparxity strategic advisor, said co-designing models, distillation pipelines, and hardware around sparsity can unlock further computing efficiency. That is the connection the new funding is intended to develop: models that choose more selectively, paired with hardware able to execute those choices.

A Commercial Base Before the Next Chip

Sparxity says its foundation includes three chip tapeouts, more than 70,000 chips produced, and 12 racks deployed. It reports more than 30 customers worldwide and revenue that quadrupled between 2024 and 2025, with continued momentum in 2026.

Its current systems address both high-performance computing and AI. The company reports 10-times-faster sparse solvers, 18-times-lower energy use for event-based machine learning, and 20-times-shorter optimization runtimes on selected workloads. These results concern particular applications on its existing systems; the up-to-44-times inference figure is a target for next-generation technology in ultra-sparse models.

The distinction is central to the financing story. Sparxity already manufactures and deploys systems, while the Series A supports a further step into hardware designed specifically for dynamic sparsity in AI inference.

Jülich Provides a Concrete Deployment Example

The Jülich Supercomputing Centre independently describes a two-rack SpiNNaker2 system from Sparxity called JUBIS, installed in April 2026. It comprises 4,320 chips and 656,640 ARM cores with neuromorphic accelerators.

JUBIS supports Jülich’s Event-based Supercomputer for AI and Beyond project, which explores real-time AI, neuroscience simulations, and other workloads such as quantum-computer emulation. The project also includes benchmarking, application prototyping, and software development to make the system useful within a broader high-performance computing environment.

That deployment illustrates why a computing architecture needs more than a chip specification. Researchers need tools, interfaces, and a way to integrate the system into their existing work. Sparxity’s full-stack approach spans that path from custom silicon to multi-rack installations.

A Route Into European Data Centers

Sparxity is also working with T-Systems on a potential route into commercial data centers. Deutsche Telekom’s November 3 announcement describes a signed letter of intent to combine Sparxity’s technology with T-Systems’ infrastructure and operating expertise.

The planned sequence starts with current hardware to develop technical integration approaches, followed by deployment of next-generation chips planned for 2028. The partners expect to agree further details over the coming months.

For Sparxity, the new capital supports a move from proving sparse computing in specialized systems toward a larger role in AI infrastructure. Its thesis is that future gains will come from doing less unnecessary work—and designing the hardware, software, and models together so those savings translate into useful performance.