To understand the along-strike variation of crustal deformation and tectonic processes in the India–Tibet continental collision zone, we deployed a linear array of broadband seismic stations along 92° E to image lithospheric structure. Our receiver-function cross-section reveals a prominent negative converter dipping ∼20° north from ∼10–55 km depth below sea-level, almost through the whole crust, beneath the southern Lhasa terrane. We interpret it to be a manifestation of the Yarlung–Zangbo suture zone (YZS) separating the continental crust of the Indian and Eurasian plates. This implies the hypothesized channel-flows of Indian middle crust extruding southwards from Tibet are limited at this longitude to the southernmost portion of the Lhasa terrane. A positive converter, consistent with previous suggestions of eclogite formation, is seen about 10–15 km above the Moho and continuing 50 km north of the 20°-dipping YZS converter. We image this positive converter continuously from ∼60 km south of the surface trace of the YZS to the vicinity of the Jiali fault, supporting the interpretation of sub-horizontal underplating of Tibetan crust by Indian crust to ∼31° N at 85° E on the Hi-CLIMB transect. However, we also show a negative mantle converter sub-parallel to the crustal YZS converter, from the northern limit of the underplating Indian lower crust to at least 140 km depth, that we interpret as the base of Tibetan lithosphere overlying an asthenospheric mantle wedge. Based on the lithospheric structure observed in this and other studies, we infer that Indian mantle lithosphere currently detaches from Indian lower crust at the “mantle suture” that is nearly 50 km south of the surface trace of the YZS at 92° E, south of the mantle suture suggested by INDEPTH transect beneath the surface trace of the YZS at ∼90° E, and far south of the mantle suture suggested to be at the 31° N northern limit of underthrusting Indian lower crust suggested by Hi-CLIMB transect at ∼85° E. A change from underplating in the west to steep subduction in the east can reconcile all these observations.