Starting with Wannier's 1) and Anderson's 2) seminal papers the triangular-lattice antiferromagnet has attracted a lot of attention. One spectacular feature of this model system for a strongly frustrated magnet is the unconventional magnetization process of the triangular-lattice Heisenberg antiferromagnet. [3][4][5][6][7][8][9][10][11][12][13][14][15][16][17] While for the classical isotropic Heisenberg model at zero temperature the magnetization M increases linearly with the applied magnetic field H, thermal or quantum fluctuations induce a plateau at 1/3 of the saturation magnetization M sat (''order from disorder'' phenomenon). For the extreme quantum case, i.e., spin quantum number s ¼ 1=2, this plateau at T ¼ 0 has been widely discussed. Very recently an almost perfect experimental realization of an s ¼ 1=2 triangular-lattice Heisenberg antiferromagnet has been reported for Ba 3 CoSb 2 O 9 , 17) where the entire measured magnetization curve MðHÞ including the 1/3 plateau is in excellent agreement with theoretical predictions. In particular, the theoretical magnetization data obtained by high-order coupled-cluster approximation 11) and by large-scale exact diagonalization approach 15) almost coincide with the measured ones over a wide range of the magnetic field.Another interesting triangular-lattice magnet is Ba 3 -NiSb 2 O 9 that is considered as a good candidate of an s ¼ 1 triangular-lattice Heisenberg antiferromagnet. 16,18,19) In Ba 3 NiSb 2 O 9 , the uniform triangular lattice of magnetic ions is realized as in Ba 3 CoSb 2 O 9 . Indeed, recently it has been found that the magnetization curve of this compound also exhibits a clear 1/3 plateau. 16) By contrast to the wellinvestigated spin-half case there are much less reliable theoretical studies of the s ¼ 1 model relevant for Ba 3 NiSb 2 O 9 . Motivated by the excellent agreement of theoretical predictions based on exact diagonalization (ED) and coupled-cluster method (CCM) and experimental results for the spin-half compound Ba 3 CoSb 2 O 9 reported in Ref. 17 we apply in this paper these methods to the s ¼ 1 model to provide theoretical data to compare with experimental results.For the s ¼ 1 model the Lanczos ED for N ¼ 27 sites and the CCM-SUBn-n approximation 20) for n ¼ 2; 4; 6; 8 are used to calculate the field-dependent properties of the model.In this short note we will not explain details of the Lanczos ED and the CCM. We refer the interested reader e.g., to Refs. 15,21, 22 and 11,[23][24][25] respectively. First we report the main theoretical results. The saturation field is H sat ¼ 9Js ¼ 9J, where J is the nearest-neighbor exchange coupling. The zero-field uniform susceptibility calculated with CCM-SUBn-n approximation is ¼ 0:10790, 0.09932, 0.09785, and 0.09679 for n ¼ 2, 4, 6, and 8, respectively. Moreover, it is useful to extrapolate the ''raw'' SUBn-n data to n ! 1 by ðnÞ ¼ c 0 þ c 1 ð1=nÞ þ c 2 ð1=nÞ 2 which yields finally our CCM estimate for the susceptibility CCM ¼ 0:0956. This value is in excellent agreement with the spin-wave res...