Researchers at Pacific Northwest National Laboratory have developed a new algorithm, Picasso, that reduces quantum data preparation time by 85%, addressing a key bottleneck in hybrid quantum-classical computing. The algorithm uses advanced graph analytics and clique partitioning to compress and organize massive datasets, making it feasible to prepare quantum inputs from problems 50 times larger than previous tools allowed. The PNNL team was able to lighten the computational load substantially by developing new graph analytics methods to group the Pauli operations, slashing the number of Pauli strings included in the calculation by about 85 percent. Altogether, the algorithm solved a problem with 2 million Pauli strings and a trillion-plus relationships in 15 minutes. Compared to other approaches, the team’s algorithm can process input from nearly 50 times as many Pauli strings, or vertices, and more than 2,400 times as many relationships, or edges. The scientists reduced the computational load through a technique known as clique partitioning. Instead of pulling along all the available data through each stage of computation, the team created a way to use a much smaller amount of the data to guide its calculations by sorting similar items into distinct groupings known as “cliques.” The goal is to sort all data into the smallest number of cliques possible and still enable accurate calculations. By combining sparsification techniques with AI-guided optimization, Picasso enables efficient scaling toward quantum systems with hundreds or thousands of qubits.