In spite of a remarkably large magnetocaloric effect at temperatures below ca. 10 K, molecular coolers are yet to become excellent magnetic refrigerants. Their main limitation is the heat transport across molecules, which is expected to decrease too drastically at cryogenic temperatures. Here, this prediction is corroborated for the [Gd2(OAc)6(H2O)4]·4H2O molecular cooler by thermal conductivity experiments and direct measurements of the magnetocaloric effect, together with numerical simulations. As a way out, a hybrid composite material is formed by attaching carboxylate-bridged Gd(III) molecules onto oxidized multi-walled carbon nanotubes. Notably, the molecular component of this composite maintains a large magnetocaloric effect, while the thermal conductivity of oriented composite buckypapers surpasses that of bulk [Gd2(OAc)6(H2O)4]·4H2O by a factor of 2–3 below 20 K. Eventually, direct measurements …