On Tuesday, Runware launched the Sonic Inference Pod, a single transportable unit designed to sit alongside hyperscalers’ massive data center projects.
The company claims the unit offers inference at a higher quality and lower cost than other serverless platforms or GPU clouds. The modular design allows capacity to be added quickly by creating new pods rather than expanding a fixed site. Flaviu Radulescu, co-founder and CEO, told TechCrunch that this approach represents the future.
“We believe distributed compute, positioned closer to end users for faster inference, is what will win in the long term,” he said. Aside from a lower price, Radulescu noted the system can scale fast, deploy anywhere there is power, and adapt quickly to new hardware. The pods use a closed-loop cooling system that can be built in days, compared to the months or years required for traditional facilities. They do not use water.
“Demand for inference is growing faster than facilities can be built,” Radulescu said. “What we want is to power the world’s intelligence, to be the backbone every AI model runs on with capacity that keeps up with demand instead of throttling it.”
Runware currently has 10 pods in deployment across the U.S., Europe, and Asia-Pacific. The company provides inference to Higgsfield AI and Wix and has 160 sites available to power its pods. Runware announced a $50 million Series A in December to provide infrastructure for companies generating images. The expansion into pods is part of the company’s core mission: providing inference to companies, rather than selling a single product.
AI labs like OpenAI and SpaceX are still racing to build data centers throughout the U.S. OpenAI is close to striking a $500 billion deal that would see it build a data center in Ohio, according to reports. Radulescu does not see those projects as a threat to the Sonic Inference Pods, describing the flexibility of the pods as a key differentiator.
“Every pod runs as part of a single network, so requests go wherever there’s capacity, closer to the users, and if one pod goes offline, traffic moves to another,” he said. A system failure means one pod is down rather than a whole fixed facility. Customers who want dedicated hardware get whole pods to themselves.
He is not too worried about other companies building this for themselves, saying simply that hardware is slow and the talent pool to build and fix the technology is small.
“A mistake in a circuit board design costs months between redesign, simulation, fabrication, testing and delivery,” he said. “Every one of those calls needs someone who understands exactly what each component does and what breaks if it’s gone.”
Building AI data centers is controversial, especially because of how many resources it uses. Communities where data centers are located have reported seeing a rise in utility costs. One day, Runware sees a world where it can run on renewable power and does not draw on the resources communities need, but that day is not necessarily today.
Radulescu said AI power use is going to increase regardless, driven by demand for inference, not by who supplies it. What Runware is focused on right now is how that demand gets met. “No transmission losses, no water in cooling, and we’re using power that already exists instead of asking for new grid capacity to be built. More inference built this way means less new grid, less water, for the same amount of compute.”
What it means
For people making things, this means new options for running AI models without waiting for massive construction projects. A company can request a pod, place it near users, and have capacity ready in days. This reduces the risk of a single point of failure and lowers the barrier to entry for those needing dedicated hardware.
