Browse Teach Curate

GroverGPT: An 8B Language Model Approximating Grover's Search Without Quantum State Simulation

A LLaMA-based 8-billion-parameter model trained on classical descriptions (circuit diagrams, QASM, natural language) of Grover search instances to predict output bit-string distributions via pattern recognition, generalizing from small to larger qubit counts and serving as a probe of the classical-simulability boundary of quantum algorithms.

B9 Text

GroverGPT: A Large Language Model with 8 Billion Parameters for Quantum Searching

GroverGPT is an 8-billion-parameter language model, built on the LLaMA architecture, trained to approximate the output of Grover's quantum search circuits from a classical description of the circuit …