Reducing CO2 emissions from road transport remains a key challenge for climate policy. This article compares specific CO2 emissions (mg/rev) in real-world urban driving conditions for a vehicle fuelled with E10 gasoline and LPG. The tests were conducted using the RDE method in Vilnius, Lithuania, using a portable exhaust gas analyser and OBD2 data. Emissions were analysed as a function of engine load and speed to standardise the comparison, using interpolated emission maps and engine operating probability distributions. The results indicate that the LPG-powered vehicle emitted an average of 8.4% less CO2 per crankshaft revolution than the E10-powered vehicle. Furthermore, LPG combustion was characterised by lower emission variability and a more uniform emission distribution, indicating a more stable combustion process. These differences are significant from the perspective of energy efficiency and the accuracy of the emission inventory under real-world conditions. The methodology enables precise comparison of the impact of fuels on emissions, regardless of driving style and travel time, and can provide a basis for developing predictive emission models for on-board systems and diagnostic tools.