Abstract Wildfire is one of the dominant disturbances in the Siberian boreal forest, affecting the capacity of forests to uptake carbon. Impacts of wildfires on forest carbon uptake are substantially varied by tree compositions, however, the loss of carbon uptake capability across different forest types remains uncertain. Here, we tracked post-fire changes in gross primary production (GPP) across five forest types in Siberia, namely, evergreen needleleaf forests (ENF), deciduous needleleaf forests (DNF), deciduous broadleaf forests (DBF), mixed forests (MF), and woody savannas (WSVN), using satellite-based observation data from 2001 to 2023. Results revealed substantial reductions in GPP after wildfire followed by gradual recovery except DBF. The GPP changes one year after wildfire (ΔGPP 1yr ) and recovery rates of GPP notably differed by forest types. In ENF, ΔGPP 1yr reached −63.4 g C m −2 yr −1 , which is larger than that in MF, DNF, and WSVN by −20.6, −31.3, and −42.7 g C m −2 yr −1 , respectively. Consistently, the GPP recovery rate is the lowest in ENF, at 63.7%. Furthermore, significant decreasing trends in 5-year mean of ΔGPP 1 yr in DNF, MF, and WSVN exceed −1.0 g C m −2 yr −1 yr −1 , indicating intensified impacts of increased wildfire intensity and/or extent on forest carbon. Our findings highlight the urgent need for establishing fire management and forest recovery strategies adapted to different forest types to enhance ecosystem resilience.