Synopsys Bold Prediction: Force Fields Will Accelerate Atomistic Simulations by 10,000× in 2026, Unlocking New Era of Discovery

For decades, atomistic simulations have powered major advances in materials science, electronics, and pharmaceuticals.

Central to these computer-based simulations is the accurate modeling of atomic interactions. To achieve this, scientists often use force fields — mathematical frameworks that define how atoms attract, repel, bind together (forming molecules or crystals), and evolve (over time and under influence). Force fields provide the “rules of engagement” for computer models by describing the forces at each point in space or the potential energy landscapes, such as electrostatic potentials surrounding charged particles, that guide molecular dynamics.

Atomistic simulations enable researchers to investigate phenomena such as chemical reactions during catalysis, protein folding pathways, and the response of materials to heat and pressure. The ability to do so virtually — before investing in costly, time-consuming physical experiments — is critical when developing and optimizing innovative materials, new drugs, and advanced nano-devices.

Up until now, however, these fields of research have been hamstrung by a fundamental bottleneck: the trade-off between accuracy and computational feasibility.

Traditional force fields, which use fixed formulas and empirical parameters to model atomic behavior, are computationally efficient and effective for simple systems. But they struggle with complex materials and chemical reactions, lacking the flexibility needed for cutting-edge discovery and innovation.

Conversely, simulations based on DFT — which model atomic interactions at the quantum level — offer far better precision but come with a steep computational cost. Simulating a few thousand atoms using traditional DFT methods can take days or weeks, and large-scale, realistic modeling involving hundreds of thousands or even millions of atoms has remained wholly impractical.

Fortunately, new types of force field models have emerged in recent years that don’t demand such trade-offs. 

The emergence and anticipated impact of MLFFs can be traced to recent hardware and software innovations. And our prediction of a 10,000× speedup in simulation workflows will be driven by four technological pillars:

1. Artificial intelligence. MLFFs learn atomic interactions from vast pre-existing datasets of quantum calculations, and they can be rapidly adapted and fine-tuned for new chemical environments using synthetic training data that is generated in real time.
2. GPU acceleration. Most MLFFs are natively designed to run on GPUs. Advances in algorithms, libraries, and hardware will continue to provide improvements in both training and simulation speeds.
3. Network-based scaling. Because GPU memory remains a limiting factor for MLFFs, high-speed interconnects that allow many GPUs to work together seamlessly will be critical — not just for boosting performance, but for scaling simulations to handle much larger system sizes.
4. Advanced simulation platforms. Platforms like Synopsys QuantumATK bring together MLFFs running on multi-GPU architectures with advanced atomistic simulation algorithms. These platforms integrate multiple computational approaches — such as DFT — within a single, flexible Python interface, making it easy for users to customize workflows, including active learning and fine-tuning of MLFFs.

The availability of these technologies and resulting democratization of advanced simulation will be a boon for research and innovation. Tasks that once required supercomputers or years of manual effort can now be completed in hours using GPU-accelerated systems. And scientists across industries can access these powerful methods through professional-grade software platforms designed for ease of use and scalability.

As more startups, academic labs, and enterprises take advantage of these capabilities, we believe atomistic simulations will become as routine in R&D as computer-aided design (CAD) became in mechanical engineering decades ago. MLFFs play a central role in this shift, enabling large-scale, high-accuracy simulations of chemical reactions, thermal transport, and other complex phenomena that were once impractical to study at this level.