The

Context. The spatial distribution of elemental abundances and their time evolution are among the major constraints to disentangling the scenarios of formation and evolution of the Galaxy. Aims. In this paper we used the sample of open clusters available in the final release of the Gaia-ESO survey to trace the Galactic radial abundance and abundance-to-iron ratio gradients, and their time evolution. Methods. We selected member stars in 62 open clusters, with ages from 0.1 to about 7 Gyr, located in the Galactic thin disc at galactocentric radii (RGC) from about 6 to 21 kpc. We analysed the shape of the resulting [Fe/H] gradient, the average gradients [El/H] and [El/Fe] combining elements belonging to four different nucleosynthesis channels, and their individual abundance and abundance ratio gradients. We also investigated the time evolution of the gradients dividing open clusters in three age bins. Results. The [Fe/H] gradient has a slope of −0.054 dex kpc−1. It can be better approximated with a two-slope shape, steeper for RGC ≤ 11.2 kpc and flatter in the outer regions. We saw different behaviours for elements belonging to different channels. For the time evolution of the gradient, we found that the youngest clusters (age < 1 Gyr) in the inner disc have lower metallicity than their older counterparts and that they outline a flatter gradient. We considered some possible explanations, including the effects of gas inflow and migration. We suggest that the most likely one may be related to a bias introduced by the standard spectroscopic analysis producing lower metallicities in the analysis of low-gravity stars. Conclusions. To delineate the shape of the ‘true’ gradient, we should most likely limit our analysis to stars with low surface gravity log g > 2.5 and microturbulent parameter ξ < 1.8 km s−1. Based on this reduced sample, we can conclude that the gradient has minimally evolved over the time-frame outlined by the open clusters, indicating a slow and stationary formation of the thin disc over the last 3 Gyr. We found a secondary role of cluster migration in shaping the gradient, with a more prominent role of migration for the oldest clusters