Fused deposition modeling (FDM) is one of the 3D printing methods that has attracted significant attention. In this method, small and complex samples with nearly no limitation in geometry can be fabricated layer by layer to form end-use parts. This work investigates the liquid sensing behavior of FDM printed flexible thermoplastic polyurethane, TPU filled with multiwalled carbon nanotubes, MWCNTs. The sensing capability of printed TPU-MWCNT was studied as a function of MWCNT content and infill density in response to different solvents, i.e., ethanol, acetone and toluene. The solvents were selected based on their widespread use and importance in medical and industrial applications. U-shaped liquid sensors with 2, 3 and 4wt.% MWCNT content were printed at three different infill densities of 50, 75 and 100%. Solvent sensitivity was then characterized by immersing the sensors in the solvents and measuring the resistance evolution over 25s. The results indicated a sensitivity order of acetone > toluene > ethanol, which was in agreement with the predictions of FloryHiggins solubility equation. For all the solvents, the sensitivity was enhanced as the infill density of the printed samples was decreased. This was attributed to the increased surface area to volume ratio and shortened diffusion paths. The MWCNT content was also observed to have a profound effect on the sensitivity; in samples with partial infill, the sensitivity was found to be inversely proportional to the MWCNT content, such that the highest resistance change was obtained for nanocomposites with the lowest MWCNT content of 2wt.%. For instance, a resistance increase of more than 10 times was obtained in 25 s once TPU-2wt.%MWCNT with 50% infill was tested against acetone. The results of this work reveals that highly sensitive liquid sensors can be fabricated with the aid of 3D printing without the need for complex processing methods.