Adsorption chillers provide sustainable cooling from waste or solar heat. However, adsorption chillers currently show limited performance. To increase the performance, new working pairs and adsorber geometries are constantly proposed. Evaluating the performance of new working pairs and adsorber geometries requires time and large amounts of the material. To reduce time and material needs, a method is presented to reliably predict the heat flows in the adsorber, specific cooling power (SCP), and coefficient of performance (COP) in an adsorption chiller from only 1 g of adsorbent material. For this purpose, the small‐scale Infrared‐Large‐Temperature‐Jump experiment is combined with a full‐scale adsorption chiller model. The adsorption chiller model allows determining time‐resolved heat flows, SCP, and COP. The prediction results are compared with a full‐scale experiment of an adsorption chiller. For various process conditions, the prediction is highly reliable with average deviations of 18.5% for the heat flows, 1.4% for the SCP, and 7.0% for the COP compared with the experiment. The presented method allows a comprehensive and reliable evaluation of new working pairs and adsorber designs from only small amounts of the adsorbent material, thus guiding material improvements at an early stage of development.