The status and prospects are reviewed for the presently most favored model
for the inner engine of cosmological gamma-ray bursts (GRBs): the black hol
e-torus model. Support for this model is derived from recent GRB-phenomenol
ogy: a cosmological origin, a bi-modal distribution in their durations, and
potentially a GRB/X-ray transient connection. The black hole-torus model o
perates by the rotation of the central Kerr black hole, which introduces tw
o new mechanisms: in-situ pair-creation on open field-lines along the axis
of rotation and a powerful magnetic black hole-to-torus coupling. The lepto
nic winds serve as input to the GRB-afterglow emissions. Long/short GRBs ar
e identified with suspended/hyper-accretion onto rapidly/slowly rotating bl
ack holes. The suspended accretion state is expected to be accompanied by b
aryonic winds blown off the torus; if formed in stellar collapse, these win
ds may account for recently detected iron line-emissions and may have contr
ibuted to the chemical abundances in X-ray transients. This implies that ME
TE-II will detect afterglows from all bursts, but iron-line emission only f
rom long bursts. In long bursts, the tents is expected to radiate most of t
he black hole luminosity in gravitational waves. This predicts that long GR
Bs are potentially the most powerful burst-sources of gravitational waves i
n the Universe. Their emissions trace a horizontal branch in the f(f)-diagr
am, which can be tested by LIGO/VIRGO. (C) 2001 Published by Elsevier Scien
ce B.V.