quantum annealing

Imagine you are trying to find the lowest point in a vast, foggy landscape of hills and valleys, where the lowest valley is the best answer to your problem. A classical search might walk downhill and get stuck in the first small dip it finds. Quantum annealing is a different style of quantum computing that tries to settle into that lowest valley by starting in a simple, easy-to-prepare quantum state and then slowly tuning the system so that its lowest-energy configuration encodes the answer you want. If you change things slowly and gently enough, the system tends to stay in its lowest-energy state the whole way, ending up in the good solution at the end. Quantum effects like tunneling can sometimes help it pass through a barrier rather than having to climb over it.

Unlike the gate model of quantum computing (the kind that runs circuits of H, CNOT, and other gates), an annealer does not run arbitrary programs. Instead, you encode an optimization problem as an energy function and let the hardware relax toward low energy. This makes it a specialized tool, well suited to certain combinatorial optimization and sampling problems, rather than a general-purpose computer. D-Wave is the company that commercialized this approach, building machines with thousands of qubits aimed at these problems.

Be careful with the hype here. Whether quantum annealing actually beats good classical algorithms (like simulated annealing or modern specialized solvers) is genuinely debated, and clear, broad speedups have been hard to demonstrate. It does not 'try all answers at once,' and it is not known to give the kind of exponential advantage that Shor's algorithm gives for factoring. Treat it as a promising, problem-specific approach whose practical edge is still being worked out, not as a proven shortcut for hard optimization in general.