Sereact gets Series B funding to expand Cortex 2.0 robot brain, enter U.S. market

Sereact gets Series B funding to expand Cortex 2.0 robot brain, enter U.S. market

Cortex 2.0 is designed to work throughout embodiments and purposes. Supply: Sereact

Sereact GmbH immediately stated it has raised Collection B funding of $110 million to scale its Cortex 2.0 “robotic mind” and assist its entry into the U.S. market. The corporate has opened an workplace in Boston and is hiring native engineering, industrial, and software staffers.

“We guess early which you can’t construct actual robotics AI in a lab. You construct it with a knowledge flywheel fed by actual deployments — transport into manufacturing, dwelling with the failures, and letting the mannequin be taught from what truly occurs on the ground,” acknowledged Dr. Ralf Gulde, co-founder and CEO of Sereact. “The numbers present it labored. 200 techniques. One billion picks. One intervention per 53,000. No one else is shut.”

Based in 2021, Sereact stated it builds bodily AI for warehouses and manufacturing. Its Cortex mind runs throughout single-arm picking cells, dual-arm returns stations, and humanoid robots, in addition to Sereact Lens, a 3D notion system for inventory and high quality management.

The Stuttgart, Germany-based company‘s current European prospects embody Daimler Truck, Mercedes-Benz, BMW, MS Direct, Energetic Ants, DeltiLog, Rohlik Group, and Austrian Submit.

Sereact begins with AI for warehouse robots

“Sereact builds AI for robots that work within the bodily world,” stated the corporate. “Warehouses had been the primary deployment as a result of no different surroundings supplies the identical mixture of knowledge factors: billions of actual interactions, each object form conceivable, arduous throughput constraints, and penalties when the robotic will get it improper.”

Sereact claimed that Cortex learns from picks of bodily objects in real-world circumstances throughout buyer websites that no simulation can reproduce. Excluding the one in 53,000 that require distant human intervention, the robotic can deal with picks by itself, feeding knowledge again into its mannequin in a closed loop.

“Each profitable choose, each failure, each restoration is captured with synchronized observations, robotic state, gripper drive suggestions, and consequence — then filtered, prioritized by novelty and uncertainty, and used to replace the mannequin,” the corporate asserted. “Up to date insurance policies move automated regression checks and roll out to the fleet. The loop closes. Information compounds. Protection of the lengthy tail expands.”

“Rivals are elevating billions to coach on simulated knowledge and lab demos,” it added. “Sereact has spent 5 years coaching on actual operations, at evening, at peak, on the messy objects that don’t appear like something the robotic has seen earlier than. It’s a niche that widens with each shift.”



Editor’s word: Bodily AI, warehouse automation, and humanoid robots can be among the many subjects mentioned on the 2026 Robotics Summit & Expo subsequent month in Boston. Register now to attend.

Cortex 2.0 anticipates outcomes earlier than it strikes

“Right now’s Cortex sees and picks,” stated Sereact. “Cortex 2.0 thinks first, then acts.”

Cortex 2.0 augments a vision-language-action (VLA) mannequin with a world mannequin. From the present state, it generates a set of potential trajectories; runs them towards a realized mannequin of physics and object conduct; and scores every one for stability, threat, and effectivity. The robotic commits to the best-scored department and updates the rollout in actual time because the scene modifications, defined the corporate.

“The shift is from try-and-see to plan-and-try,” Sereact stated. “Right now’s reactive insurance policies, once they miss, are inclined to repeat the identical movement and compound the failure.”

“Cortex 2.0 evaluates a number of outcomes first and guidelines out the unhealthy ones earlier than the arm strikes — which is precisely what’s wanted for the form of work the place contact issues: assembling a part underneath stress, inserting a windshield wiper with out scratching it, kitting elements that must land in precisely the proper orientation for the following station,” it added.

Cortex 2.0 plans in “visible latent area,” in response to Sereact. Whereas joint instructions are tied to a selected robotic’s kinematics, the pixels encode regularities about objects, contact, and movement that switch throughout embodiments.

“The robotic desires in latent area,” stated Marc Tuscher, co-founder and chief know-how officer of Sereact. “We give it a type of creativeness — the power to anticipate how the world will reply earlier than it strikes.”

“We don’t construct robots. We don’t promote companies. We ship one factor: the mannequin that runs on any robotic,” he stated. “Single arms, twin arms, humanoids, fastened cells — similar mind throughout all of it. {Hardware} is changing into a commodity. The mannequin isn’t.”

Sereact additionally famous that compute planning might be tuned to the duty. The know-how can present extra foresight when failure is dear, reminiscent of in parcel packing, kitting, or inserting fragile objects. It might plan much less when restoration is affordable, like for a regrasp on a missed choose. “Cortex 2.0 spends planning finances the place it pays again,” it stated.

Sereact is developing models that can run across different robots and applications.Cortex is designed to be taught from knowledge from real-world experiences. Supply: Sereact

Companies spend money on the info flywheel

Headline led Sereact’s newest funding, which included participation from Bullhound Capital, Daphni, and Felix Capital. Current traders Air Road Capital, Creandum (lead of Sereact’s 2025 Series A), and Level 9 all returned for the Collection B spherical. The corporate has raised over $140 million up to now.

“The bodily AI alternative is among the largest we’ve seen in a technology, and we imagine it should rewire world provide chains and manufacturing,” stated Trevor Neff, development accomplice at Headline.

“Behind nice alternatives and nice corporations are nice founders, and Ralf and Marc are constructing into that chance the proper manner: actual deployments, actual knowledge, and a mannequin that compounds and will get higher with each single choose,” he added. “Clients love the product, which results in continued growth, solely accelerating the info flywheel — that is why we’re so excited to again Sereact.”

“After a deluge of humanoid robotics corporations, my fellow Companion Alon Kuperman and I had been delighted to satisfy Ralf and Marc, the co-founders of Sereact, who’ve constructed an AI working system that seamlessly retrofits into the world’s huge fleet of business robots already in motion,” stated Per Roman, founding accomplice at Bullhound Capital.

Antoine Nussenbaum, co-founder and investor at Felix Capital stated: “We see Sereact as a brand new technology European champion, combining deep industrial know-how with world-class AI and robotics expertise to sort out one of many hardest challenges in world provide chains. At its core is a strong, compounding knowledge moat, the place each warehouse interplay makes the system smarter and more durable to duplicate.”

The submit Sereact will get Collection B funding to develop Cortex 2.0 robotic mind, enter U.S. market appeared first on The Robotic Report.