The Hi-5 instrument, a proposed high-contrast L' band (3.5-4.0 μm) nulling interferometer for the visitor focus of the Very Large Telescope Interferometer (VLTI), will characterize young extra-solar planetary systems and exozodiacal dust around nearby main-sequence stars. Thanks to VLTI's angular resolution (λ=B = 5 mas for the longest UT baseline), it will fill the gap between young giant exoplanets discovered by ongoing single-aperture direct imaging surveys and exoplanet populations discovered by radial velocity surveys. In this paper, we investigate the exoplanet detection yield of Hi-5. First, we present the latest catalog of stars identified as members of young stellar associations within 150 pc of the Sun thanks to the BANYAN algorithm and other searches for young moving group members. Realistic exoplanet populations are then generated around these stars and processed with the SCIFYsim tool, the end-to-end simulator for the Hi-5 instrument. Then, two formation models are used to estimate the giant planet's luminosity. The first is the New Generation Planetary Population Synthesis (NGPPS), also known as the Bern model, and the second is a statistical model based on gravitational instability (hot-start model - AMES-Dusty model). We show that Hi-5 is insensitive to cold-start planets but can detect giant hot-start planets. With ATs, more than 40 planets could be detected assuming 20 nights of observations. With its unique capabilities, Hi-5 is also able to constrain in mass the observed systems. Hi-5 is sensitive to planets with a mass > 2 Mjup around the snow line.