All spaceborne microwave imaging and positioning techniques are affected by the tropospheric delay. The troposphere is non-dispersive at frequencies investigated in this study, thus, the delays estimated by one technique can be useful for mutual corrections. We compare the tropospheric delays retrieved from the Global Navigation Satellite Systems (GNSS) with multi-baseline spaceborne Synthetic Aperture Radar Interferograms (InSAR). One of the limitations of the InSAR technique are the atmospheric effects that may be misinterpreted as deformations. The InSAR satellites flights are separated by days or even weeks. During this time, the tropospheric conditions, especially the water vapor density, may change completely. The impact of the troposphere on the InSAR images is usually eliminated using two approaches: estimation of the atmospheric phase corrections from the high- resolution interferometric phase itself (suitable in particular if many acquisitions are available) or introducing high-resolution external troposphere models (especially when only a few acquisitions are available). In this study, we present high-resolution models of differenced slant tropospheric delays (dSTD) for alpine regions based on GNSS data. The dSTD models in this study are double-differenced slant delays w.r.t. the reference point and the reference (master) SAR acquisition. The models are interpolated to the locations of InSAR points using the in-house developed software package COMEDIE (Collocation of Meteorological Data for Interpretation and Estimation of Tropospheric Pathdelays). The models are calculated for 32 acquisitions of COSMO-SkyMed satellite in a period between 2008 and 2013. The acquisitions were taken from the snow-free period from June to October, always around similar time, some minutes before 6 pm. The chosen research area of around 20 km x 25 km is located in Zermatt and Matter Valley in the Swiss Alps. The GNSS-based models are calculated from 5 to 8 stations located within or close to the area of the study for the years 2008 – 2011 and from 11 stations for years 2012 – 2013. We compare the GNSS and InSAR-derived tropospheric delays. The preliminary results show a good agreement between InSAR and GNSS estimates for some of the interferometric layers. The correlation coefficient averaged from all of the interferograms is equal to 0.64. The average bias of the residuals InSAR – GNSS equals -6 mm with a standard deviation of 4 mm. For the period of only 2012 – 2013 with more GNSS data, the bias is reduced to -2 mm. We are currently working on adding also the information from the low-cost L-1 only GPS stations located within the area of the study. In the next steps, the GNSS delays will be introduced as a priori models at the interferogram level and the reasons for the poor agreement between InSAR and GNSS estimates for some of the layers will be further investigated.