The present work resulted in successful demonstration of vibrational up-pumping of nitric oxide with a line-selected CO laser both at steady state and at repetitively pulsed conditions. The laser was operated on a 8→7 P(11) line which has a very good resonance with the 0→1 vibrational transition of NO, with output power of approximately 5W. First overtone infrared emission from vibrational levels up to v=14 was measured with a Fourier Transform Infrared spectrometer operating in both rapid-scan (for steady-state measurements) and time-resolved step-scan (for time-resolved measurements) modes for three different NO partial pressures, P=0.1, 0.2, and 0.3 Torr, and Ar diluent pressure of PAr=100 Torr. The experimental results demonstrated that up to 14 NO vibrational levels (at steady state) and up to at least 10 vibrational levels (in time-resolved measurements) are populated and radiating. Vibrational distribution functions inferred from the steady-state IR emission spectra demonstrate deviation from the Boltzmann distribution, due to the effect of an-harmonic vibration-vibration pumping. A kinetic model, incorporating processes of laser beam absorption and stimulated emission, vibration-vibration (V-V) and vibration-translation (V-T) energy exchange, spontaneous radiative decay of NO, and diffusion of vibrationally excited molecules out of the laser-excited volume was used to model the experimental results. Comparison of the experimental results and the modeling calculations showed that the predicted time-resolved NO IR emission lags behind the experimentally measured emission, both during the excitation and during the relaxation. Incorporating Gaussian laser beam power distribution improved agreement for low vibrational levels but varying the NO-NO V-V rates, incorporating multi-quantum V-V and vibrational energy “sink” processes, and varying the focused laser beam diameter and the emission signal collection pattern did not result in further improvement of the agreement with the experiment. Comparative analysis of previous and the present optical pumping experiments strongly suggests that a combination of two factors, (1) the use of a focused pump laser beam, and (2) line-of-sight averaging signal collection introduces significant uncertainty into the input parameters of the kinetic model. For this reason, accurate kinetic modeling and inference of the V-V rates from these experiments is difficult.