We prove that the energy gap of the model proposed by Zhang et al (2016 arXiv:1606.07795) is exponentially small in the square of the system size. In Movassagh and Shor (2016 Proc. Natl Acad. Sci. USA) a class of exactly solvable quantum spin chain models was proposed that have integer spins (s), with a nearest neighbors Hamiltonian, and a unique ground state. The ground state can be seen as a uniform superposition of all s-colored Motzkin walks. The half-chain entanglement entropy provably violates the area law by a square root factor in the system's size (∼√n) for s > 1. For s = 1, the violation is logarithmic (Bravyi et al 2012 Phys. Rev. Lett. 109 207202). Moreover in Movassagh and Shor (2016 Proc. Natl Acad. Sci. USA) it was proved that the gap vanishes polynomially and is O(n-c) with c ≥ 2. Recently, a deformation of Movassagh and Shor (2016 Proc. Natl Acad. Sci. USA), which we call 'weighted Motzkin quantum spin chain' was proposed Zhang et al (2016 arXiv:1606.07795). This model has a unique ground state that is a superposition of the s-colored Motzkin walks weighted by with t > 1. The most surprising feature of this model is that it violates the area law by a factor of n. Here we prove that the gap of this model is upper bounded by 8nst-n2/3 for t > 1 and s > 1.