Abell 2744-QSO1 combines three characteristics rarely observed in a single system. Its light comes from an epoch situated roughly 700 million years after the Big Bang, its black hole reaches nearly 50 million solar masses, and the stellar mass of its host galaxy remains lower than that of the black hole. The dynamical measurements published in 2026 indicate a black-hole-to-stellar-mass ratio greater than two [1].
This disproportion compels us to revise an overly simple chronology of galaxy formation. Classical scenarios have black-hole seeds born from the first stars, then grown by accretion and merger within galaxies already engaged in their evolution. QSO1 shows that a massive nucleus was able to take a considerable lead over the stellar component surrounding it. The reach of the observation remains confined to the history of the first black holes and their hosts, without calling the cosmological framework as a whole into question.
An object from cosmic infancy
QSO1 lies at a redshift of z = 7.04. The object appears as three images behind the cluster Abell 2744, whose mass curves space-time and amplifies the light coming from the background. This gravitational lens, combined with observations from the NIRSpec spectrograph on the James Webb telescope, made it possible to study a system that would otherwise remain too compact and too faint. The first spectra had already identified a strongly reddened active nucleus and provided a virial estimate of its mass [2].
The object belongs to the Little Red Dots, a population of compact red sources discovered in numbers by James Webb. This designation covers different systems. Some spectra show broad lines associated with an accreting black hole, others are more compatible with compact star formation, while several objects could combine both components [3]. QSO1 stands out through the dominance of its nucleus and the weakness of the detectable stellar component.
A seed issuing from the collapse of a massive star would have had to grow very rapidly to reach the observed mass in the time available. Such an evolution requires a durable supply of gas, efficient accretion and conditions able to sustain the feeding of the nucleus. The low stellar mass of the host makes this trajectory harder to reconstruct, without excluding it on its own.
A particularly precious dynamical measurement
The mass of very distant black holes is generally estimated from the luminosity and the width of the emission lines. These virial relations are calibrated in the nearer Universe and their application to Little Red Dots remained debated. For QSO1, the gravitational lens and deep spectroscopic data made possible a measurement based on the motion of the gas around the centre.
The velocity field of the narrow Hα line is compatible with Keplerian rotation about a point mass. Three-dimensional modelling, corrected for the estimated inclination, leads to a mass of about 5 × 107 solar masses. An extended stellar distribution reproduces the data less well and tends, in the fits, to contract toward an almost point-like mass. The dynamical estimates thus agree with the order of magnitude previously obtained by virial methods [1].
The measurement retains uncertainties bound up with inclination, spatial resolution, the structure of the gas and the lens model. The authors also tested contamination by outflows and several mass profiles. These limits affect the precision of the result, while the preference for a compact centre remains above five standard deviations in the published kinematic analysis. The best-supported interpretation remains that of a supermassive black hole whose mass exceeds that of the stars detectable in the host [1].
A disproportion that overturns intuition
In nearby galaxies, the mass of the central black hole usually represents a small fraction of the stellar mass. QSO1 lies several orders of magnitude above these local relations, with an upper limit of about 2 × 107 solar masses for the stellar component. The system documents a highly unbalanced state in which the nucleus has outpaced the visible growth of its host [1].
The surrounding gas also shows a metallicity below one hundredth of the solar value, judging by the weakness of the [O III] line relative to Hβ [4]. Such limited chemical enrichment indicates that few generations of stars have had time to produce and then disperse heavy elements. The coexistence of a massive black hole and an almost pristine medium constrains scenarios that rest on a long succession of stellar episodes.
This head start may alter the later evolution of the galaxy. The gravity of the nucleus concentrates gas, while its radiation and its winds can heat or expel it. The net effect depends on the geometry, on the density of the medium, on the accretion rate and on the channels for evacuating energy. On a systemic reading, QSO1 illustrates an early centre able to influence the formation conditions of its own environment, without thereby becoming the sole cause of the galaxy.
The possible formation scenarios
Direct collapse. A primordial cloud can avoid fragmenting into stars if molecular cooling remains limited. In a sufficiently massive halo, the gas then concentrates toward the centre and forms a heavy seed, of the order of 104 to 106 solar masses. This initial mass strongly reduces the growth needed to reach that of QSO1 [5]. The scenario nevertheless requires a medium very poor in metals and, in its classical versions, an intense ultraviolet source able to dissociate molecular hydrogen. No source of this kind is clearly observed near QSO1 [1].
Stellar seeds and rapid accretion. The first stars, known as Population III, may have produced black holes heavier than ordinary stellar remnants. Accretion episodes above the Eddington limit would then accelerate their growth. The black hole of QSO1 is currently accreting at a low rate relative to that limit, which does not exclude much more active earlier episodes [1]. This route depends on a reservoir of dense gas and on a mechanism able to feed the centre despite the radiation it emits.
Primordial black holes. Another family of models invokes objects formed by the collapse of density fluctuations in the very young Universe, before the first stars. QSO1 provides no independent proof of their existence. Its low metallicity and its extreme ratio of black-hole to stellar mass nevertheless make this hypothesis compatible with the observed properties. Simulations devoted to this case show that a massive primordial seed could delay star formation through its energetic feedback, but this result remains dependent on the model and on parameters that are still poorly constrained [6].
No scenario reproduces all the characteristics of the system without difficulty. A combination remains possible, for example a heavy seed followed by episodes of rapid accretion in a gas-rich environment. Present observations establish the mass, the imbalance with the host and the low metallicity with differing degrees of confidence. They do not yet allow a single trajectory to be reconstructed.
Little Red Dots: a multiple population
The significance of QSO1 also depends on the population to which it belongs. Little Red Dots represent a substantial share of the broad-line active nuclei detected at high redshift, but their selection by colour groups together objects of different nature and evolutionary stage [1][3]. Models of stellar atmospheres, dense gas and active nuclei can sometimes produce similar signatures, which makes spectroscopic confirmation necessary for each source.
QSO1 offers a particularly solid case because the variability of the lines, the broad and narrow components of the spectrum, and then the kinematics of the gas all converge toward the presence of an accreting black hole [1][7]. This convergence cannot be extended automatically to the whole population. It does demonstrate, however, that at least some Little Red Dots harbour massive nuclei well before the stabilisation of the relations observed in nearby galaxies.
Reionisation: a further actor in the young Universe
Accreting black holes emit ultraviolet and X-ray radiation capable of ionising the intergalactic gas. Their contribution to reionisation depends on their abundance, on their luminosity and above all on the fraction of radiation able to escape the dense envelopes that surround many Little Red Dots. The red colour of these objects often signals strong absorption, which limits any extrapolation from their intrinsic luminosity.
Deep ultraviolet observations of QSO1 have brought to light a broad Lyα profile and a nearby environment containing associated sources. The authors of that study propose that Little Red Dots may be found preferentially in regions already ionised by their dense surroundings [8]. This result concerns the local conditions around two objects and does not yet measure the global contribution of this population to reionisation.
Galaxy–black-hole relations still immature
The correlations observed today between the mass of black holes and that of galactic bulges may represent the outcome of billions of years of accretion, star formation, mergers and feedback. QSO1 gives access to an earlier, highly dispersed phase in which nucleus and host have not yet converged toward these relations. Other overmassive objects discovered by James Webb indicate that this imbalance is not confined to a single case, even if QSO1 occupies an extreme position [1].
Distinguishing between the formation routes will require several combined diagnostics: gas metallicity, stellar mass of the host, disc geometry, variability, spectroscopic depth and statistics over a larger sample. Future gravitational-wave observations could complete these data. Pulsar timing arrays probe above all mergers of far more massive black holes, while a space detector such as LISA would be suited to intermediate seeds and their mergers.
Scope and limits of the observation
QSO1 demonstrates neither that all supermassive black holes preceded their galaxy, nor that primordial black holes exist. Routes based on stellar seeds, direct collapse and phases of rapid accretion all remain open. The phrase « black hole formed before its galaxy » summarises a primacy of growth inferred from the observed mass ratio. It does not allow the birth of the black hole, the halo, the gas and the first stars to be dated separately.
The most directly supported hypothesis concerns the state of the system at the moment observed: a black hole of about 50 million solar masses dominates a galaxy whose stellar mass is below 20 million solar masses, in a weakly enriched medium. The mechanism that produced this configuration remains undetermined. This separation between measurement and interpretation avoids turning an exceptional object into general proof about the origin of black holes.
A chronology to be revised
QSO1 establishes that a supermassive black hole could already dominate its host roughly 700 million years after the Big Bang. The dynamical measurement reinforces the earlier virial estimates and reduces the possibility of explaining the object by a compact star cluster. The low stellar mass and the very low metallicity tighten the constraints imposed on models still further.
The chronology of the first galaxies must incorporate trajectories in which the growth of the nucleus largely precedes that of the stars. The order of appearance then modifies the gas flows, the star formation and the feedbacks that shape the system. The available data describe this asymmetry with new precision, while the origin of the seed remains an open question.
Principal references
[1] Juodžbalis, I. et al. « A direct black-hole mass measurement in a little red dot at high redshift ». Nature 653, 1017-1021 (2026). DOI
[2] Furtak, L. J. et al. « A high black-hole-to-host mass ratio in a lensed AGN in the early Universe ». Nature 628, 57-61 (2024). DOI
[3] Matthee, J. et al. « Little Red Dots: An Abundant Population of Faint Active Galactic Nuclei at z ≈ 5 Revealed by the EIGER and FRESCO JWST Surveys ». The Astrophysical Journal 963, 129 (2024). DOI
[4] Maiolino, R. et al. « A black hole in a near-pristine galaxy 700 million years after the Big Bang ». Monthly Notices of the Royal Astronomical Society 548, 1-19 (2026). DOI
[5] Bromm, V. and Loeb, A. « Formation of the First Supermassive Black Holes ». The Astrophysical Journal 596, 34-46 (2003). DOI
[6] Zhang, S., Liu, B., Bromm, V. and Kühnel, F. « Primordial Black Holes as Seeds for Extremely Overmassive AGN Observed by JWST ». Preprint (2026). arXiv
[7] D’Eugenio, F. et al. « BlackTHUNDER strikes twice: Balmer-line absorption in an overmassive Little Red Dot at z = 7.04 ». Monthly Notices of the Royal Astronomical Society 547 (2026). DOI
[8] Tang, M. et al. « SPURS: Evidence for Clumpy Neutral Envelopes and Ionized IGM Surrounding Little Red Dots in Abell 2744 from Ultra-Deep Rest-UV Spectroscopy ». Preprint (2026). arXiv