The existing human activity suitability evaluation of land resources in China has the defects of insufficient theoretical support in terms of basic data accuracy, elevation and slope classification and suitability grade judgement. When it is applied empirically, there are some problems such as excessive evaluation factors, difficulty in obtaining indicator data, and low altitude of applicable regions. In this paper, we constructed a technical evaluation framework applicable to the Qinghai-Tibet Plateau, and established a grid-scale evaluation model for construction land suitability and cultivated land suitability after selecting key parameters that vary with topographic elevation and slope. And then we proposed a generalized algorithm for key parameters such as atmosphere density, air temperature, slope suitability for construction, and soil erosion resistance of sloping cultivated land. With Milin County in southeastern Tibet as an example, the empirical research has been carried out according to the interval of 100 m in elevation and 1° in slope. In addition, grid precisions of 30 m, 50 m, 100 m, 250 m, 500 m, and 1000 m are tested. This article obtains four conclusions: (1) The construction land suitability and cultivated land suitability can be classified into five levels: suitable, relatively suitable, moderately suitable, less suitable, and unsuitable. Under the grid precision of 30 m, the five levels of construction land suitability account for 1.26%, 2.92%, 2.12%, 2.77%, and 90.93%, respectively. Compared with 30 m, the area ratio results of 50 m, 100 m and 250 m have the errors of ±0.34%, ±0.34% and ±1.11%, respectively. Similarly, the area ratio of each level of cultivated land suitability with a 30 m grid precision is 0.72%, 3.07%, 2.62%, 1.35%, and 92.25%, the errors between it and the results under grid precision of 50 m, 100 m, and 250 m are ±0.52%, and the errors of 500 m and 1000 m grid precisions for both construction land suitability and cultivated land suitability are relatively larger than those of the other grid precisions. (2) Current construction lands (i.e., settlements of cities, towns and villages, industrial and mining lands) in the study area are mainly distributed in areas with levels of relatively suitable, suitable, and moderately suitable according to the distribution ratio. Under grid precisions of 30 m, 50 m, and 100 m, the total land areas of these three levels account for 95.96%, 95.96%, and 94.74%, respectively. Current cultivated land and plantation land are distributed in areas with levels of relatively suitable, moderately suitable, and suitable, accounting for more than 96% of different grid precisions. (3) The empirical research shows that evaluation method, quantitative model, and parameter algorithm for evaluating the human activity suitability of land resources in the Qinghai-Tibet Plateau are feasible and applicable. In the actual evaluation, the grid precision should be controlled within 100 m, and the maximum cannot exceed 250 m. (4) This study established a corresponding pedigree between construction land suitability and cultivated land suitability and topographic elevation and slope that can be applied to the Qinghai-Tibet Plateau. Meanwhile, some professional defects are analyzed of available land resources evaluation methods in major function-oriented zoning and "Double Evaluations" of territorial spatial planning in China when applied to the Qinghai-Tibet Plateau.