Embedl is proud to announce that we have once again been selected for Ny Teknik’s 33-listan, marking the second consecutive year that our company has been recognized among Sweden’s most innovative and promising deeptech startups. December 3, we will participate in the 33-listan 2025 finale, ' A Day on Innovation and Tech Leadership,' hosted by Ny Teknik at Clarion Hotel Stockholm.
What Is NyTeknik’s 33 List?
NyTeknik’s 33 list highlights Sweden’s top tech startups. Companies on the list are recognized as innovators with potential for substantial impact. Being featured twice demonstrates that Embedl consistently develops technologies with practical relevance.
Background of Embedl
Embedl emerged from research conducted at Chalmers University of Technology. The company optimizes AI model efficiency by reducing computational needs, enabling AI to run on smaller devices. The company holds patents for solutions that make AI leaner and faster without losing precision.
What Does “Brain Surgery” on AI Models Mean?
Embedl calls its optimization process “brain surgery” for AI models. Like a surgeon removing only what’s unnecessary, Embedl prunes models to maintain performance while reducing size and complexity.
The Technical Process
Embedl compresses neural networks by removing parts that contribute little to outcomes. The goal is to maintain accuracy while speeding up models and reducing energy consumption.
Energy Reduction Focus
AI models often use significant energy, especially in data centers and embedded systems. Embedl’s work cuts energy use by 20% to 95%, depending on the case, lowering both costs and environmental impact.
Improvements in AI Performance
Besides reducing energy consumption, Embedl’s methods accelerate AI models by two to five times. Faster processing is crucial for real-time systems, such as autonomous vehicles.
Practical Uses
Embedl’s technology is already used by companies in several industries:
- In autonomous driving, their optimizations enable the processing of sensor data more efficiently.
- In eye tracking, companies like Tobii use Embedl’s solutions for smoother performance.
- In the defense sector, Embedl’s technology supports efficient AI in mission-critical applications.
The Role of Efficiency in AI’s Future
As AI models become larger and more complex, the demand for efficient processing grows. Embedl’s focus on optimization addresses this need by reducing energy consumption and improving speed without sacrificing quality. This approach contributes to the development of more sustainable AI solutions.
Meet Us in Stockholm on 3 December
We look forward to joining innovators, industry leaders, and fellow deep tech companies at the Clarion Hotel Stockholm for one of the most important technology events of the year. If you plan to attend, we welcome the opportunity to connect and discuss the future of energy-efficient AI.