Organic acids are important components of dry tobacco and significantly affect its sensory properties. Therefore， determining the organic-acid composition and content in tobacco is greatly significant for evaluating its quality. Organic acids in tobacco are classified according to volatility as either volatile， semi-volatile， or non-volatile， with semi-volatile and non-volatile organic acids typically methylated for identification purposes. However， volatile acids cannot be determined in this manner owing to their low molecular weights. Therefore， a comprehensive， efficient， and accurate method for rapidly and simultaneously determining volatile， semi-volatile， and non-volatile organic acids in tobacco needs to be established. Herein， an innovative gas-chromatography–mass-spectrometry-based （GC-MS-based） isopropyl-esterification method was developed for simultaneously determining 29 volatile， semi-volatile， and non-volatile organic acids in tobacco. The developed method provides esterified small-molecule acids of lower volatility than those obtained using classical methylation and avoids interference from byproducts obtained via the esterification reaction. Experimental conditions and GC parameters were optimized， which ultimately led to 10% sulfuric acid in isopropanol （containing 0.1 g/mL p-toluenesulfonic acid） as the esterification reagent， with the reaction performed for 3 h at 70 °C and 500 r/min. The mixture was then cooled to room temperature， 10 mL of deionized water and 2 mL of n-hexane were added， and the mixture was vortexed for 30 s （4×）， after which the supernatant was collected for analysis. The DB-FASTFAME （30 m×0.25 mm×0.25 μm） and DB-5MS （30 m×0.25 mm×0.25 μm） chromatographic columns were compared， with the former finally selected for GC-MS separation and quantitative detection. All method-validation experiments were subsequently performed under the established instrumental and derivatization conditions. The 29 organic acids exhibited good linearities within the detection range， with all correlation coefficients （r²） exceeding 0.99. All recoveries at three spiked levels ranged from 81.8% to 118.7%. The intra-day repeatabilities （RSD， n=6） ranged from 0.5% to 8.1%， and the inter-day repeatabilities （RSD， n=6） ranged from 0.9% to 8.2%. The limits of detection （S/N=3） were 0.01-49.02 μg/g， and the limits of quantification （S/N=10） were 0.03-163.41 μg/g， respectively. Comparative experimental results revealed that the developed method is highly consistent with the classical methylation-based derivatization method for the detection of semi- and non-volatile organic acids in tobacco. Furthermore， the new method is able to simultaneously determine volatile organic acids that cannot be analyzed using the classical methylation method. Therefore， the developed method is more widely applicable for determining acids in tobacco. The newly developed method was used to determine three types of tobacco sample： flue-cured tobacco， cigar tobacco， and burley tobacco， with all 29 organic acids successfully detected. Cigar tobacco leaves were found to contain significantly more volatile organic acids than flue-cured tobacco， especially 3-methylvaleric acid； they also contain less semi-volatile organic acids and more non-volatile organic acids than flue-cured tobacco. The developed method is simple， accurate， and suitable for the large-scale simultaneous determination of volatile， semi-volatile， and non-volatile organic acids in tobacco； it also provides additional technical support for the study of organic acids in tobacco and evaluating tobacco quality.