Information is physical -- it is stored and processed in physical systems. So, the laws of information processing -- e.g., what is possible -- are derived from the physics of the underlying bits. Bits that obey quantum physics (a.k.a. "qubits") enable a whole new set of protocols and algorithms, collectively known as quantum information science. After ~30 years of effort, we have: (1) a variety of applications for quantum computers; and (2) several promising experimental technologies for building quantum computers. Unfortunately, most of the applications require thousands of qubits… and quantum devices of this size are probably 10-20 years away. Fortunately, there ARE interesting and useful applications for quantum information processors comprising 100, 10, or even just 2 qubits! Such devices aren't useful for computation -- they can be simulated effectively on classical computers -- but they can process existing quantum information with remarkable efficiency. I'll present two such examples. First, quantum devices as small as 1 qubit can be used to enhance the detection of small forces, by discriminating with optimal accuracy between distinct quantum states. Second, somewhat larger quantum devices can be used to compress quantum data, and to concentrate quantum entanglement -- which can then be used as a resource to enable large-scale quantum computing.