The uncertainty of pesticide residues in vegetables is an important indicator to judge whether the detection of pesticide residues is good or not in quality, which is too difficult to evaluate if the distribution function of the detected value is unknown. This study proposes an evaluation method based on the gray model to solve this problem. Steps are adopted as follows: first of all, detect the nine sorts of organochlorine and pyrethroid pesticide residues in vegetables by use of gas chromatography; secondly, establish, taking the detection of pesticide residues as a kind of gray process, the mathematical model of gray evaluation on the uncertainty of pesticide residues in vegetables; thirdly, use the model to evaluate the uncertainty of the above nine pesticide residues; and finally, make a comparison of the statistical result in the residues with that of the evaluation on the uncertainty. It is indicated that the evaluation by gray model is faster to calculate the uncertainty of the residues, and the correlation between those two results is 0.997 with high relativity if compared with the calculation value of the statistical standard deviation. And it is concluded that the method of evaluation is suitable for the detection of small samples and unknown distribution types with the advantages of fast, accurate and reliable in calculation, which has good effect in the application.