The objective of this study is the assessment of the real-world environmental performance, and its comparison with laboratory measurements, of two Euro 6 passenger cars. The first is equipped with a common-rail diesel engine, Lean NO x Trap (LNT), and Diesel Particulate Filter (DPF), and the second is a bi-fuel gasoline/CNG (Compressed Natural Gas) vehicle equipped with a Three-Way Catalyst (TWC). The experimental campaign consisted of on-road and chassis dynamometer measurements. In the former test set, two driving routes were followed, one complying with Real Driving Emissions (RDE) regulation and another characterized by more dynamic driving. The aim of the latter route was to go beyond the regulatory limits and cover a wider range of real-world conditions and engine operating areas. In the laboratory, the WLTC (Worldwide harmonized Light vehicles Test Cycle) was used, applying the real-world road load of the vehicles. Both cars underwent the same tests, and these were repeated for the primary (CNG) and the secondary (gasoline) fuel of the bi-fuel vehicle. In all of the tests, CO 2 and NO x emissions were measured with a Portable Emissions Measurement System (PEMS). The results were analyzed on two levels, the aggregated and the instantaneous, in order to highlight the different emissions attributes under varying driving conditions. The application of realistic road load in the WLTC limited its difference from the RDE-compliant route in terms of CO 2 emissions. However, the aggressive driver behavior and the uphill roads of the Dynamic driving schedule resulted in approximately double the CO 2 emissions for both cars. The potential of natural gas to reduce CO 2 emissions was also highlighted. Concerning the NO x emissions of the diesel car, the real-world results were significantly higher than the respective WLTC levels. On the other hand, the bi-fuel car exhibited very low NO x emissions with both fuels. Natural gas resulted in increased NO x emissions compared to gasoline, always remaining below the Euro 6 limit, with the only exception being the Dynamic driving schedule. Finally, it was found that the overall cycle dynamics are not sufficient for the complete assessment of transient emissions, and the instantaneous engine, and aftertreatment behavior can reveal additional details.