Every energetic supernova of this type that astronomers had confirmed came paired with some sign of a jet moving near light speed — a gamma-ray burst, or at least its afterglow — until a satellite caught only the second such explosion ever recorded in the act, 500 million light-years away, and it arrived with neither.

A supernova’s shock breakout, the instant the explosion’s shockwave finally punches through the dying star’s outer surface and floods space with the first light of the blast, lasts anywhere from seconds to a few hours. Theory says it happens in every core-collapse supernova. In practice, astronomers had confidently caught it only once before, in a star called SN 2008D, back in 2008. Everything else has been either missed entirely or reconstructed after the fact from what the explosion left behind. “These events have historically been difficult to detect because they are brief and because this part of the transient sky has lacked sensitive, wide-field monitoring,” Greg Zeimann of the McDonald Observatory said of the broader challenge.

On 21 March 2026, China’s Einstein Probe, an X-ray satellite built with the European Space Agency and Germany’s Max Planck Institute for Extraterrestrial Physics, caught a second one: a fast X-ray flash designated EP260321a, coming from a galaxy roughly 500 million light-years away. Follow-up observations confirmed it as the shock breakout of a broad-lined Type Ic supernova, now catalogued as SN 2026gzf. The results, from two research teams led by Brendan O’Connor at Carnegie Mellon University and Jillian Rastinejad at the University of Maryland, were published in July in The Astrophysical Journal Letters.

What should have come with it, and didn’t

Broad-lined Type Ic supernovae, the specific flavour of stellar death that produced SN 2026gzf, are the kind astronomers most associate with long gamma-ray bursts, the enormously energetic flashes produced when a dying massive star launches a jet of matter travelling at a meaningful fraction of the speed of light. Every previous supernova of this class energetic enough to be considered a strong GRB candidate has come packaged with some sign of that jet: a burst of gamma rays, or at least an afterglow across other wavelengths once the burst itself faded.

SN 2026gzf didn’t have either. “SN 2026gzf looks remarkably similar to other energetic supernovae that have been previously linked to gamma-ray bursts,” O’Connor said, according to a EurekAlert release on the findings. “Yet multi-wavelength follow-up observations using the most sensitive facilities found no evidence for a relativistic jet or an afterglow, which are typically seen in those events.” The X-ray shock breakout itself was also unusually faint, the faintest yet recorded for this class of supernova, going by the O’Connor team’s paper posted to arXiv.

A jet that never made it out

The explanation the two teams favour is that a jet did form, but never escaped the star. In the paper’s terms, the outflow was “choked,” either smothered by the dying star’s own outer layers or by a dense shell of material the star had shed in the lead-up to its collapse, before it could break through and produce a visible gamma-ray burst. The physics places real limits on what that choked jet could have looked like. The paper constrains its Lorentz factor, the standard measure of how relativistic an outflow is, to below 30, well under the values of a hundred or more expected of a jet that successfully breaks out and produces a gamma-ray burst, and its kinetic energy to under 10⁴⁹ erg, modest by the standards of a genuine GRB jet. Those numbers describe something that started moving and lost, rather than something that never tried.

Rastinejad’s team focused on what that choking material actually was. “Our observations allowed us to study the physics of three pieces of this explosion: the X-ray shock breakout, the accompanying supernova, and the interaction of the supernova with material previously cast out by the dying star,” she said, in the same release. “With this information we were able to map out the structure of the material surrounding the star and understand the star’s violent lifestyle before it collapsed.” The progenitor appears to have been a Wolf-Rayet star, a class of massive star that has already shed its outer hydrogen and helium before it dies, and one that shed mass in an irregular, clumpy way rather than a single clean shell.

Catching it required five observatories working together

None of this came from a single instrument. Einstein Probe’s detection triggered a rapid response across the Dark Energy Camera on the Víctor M. Blanco 4-metre telescope in Chile, the Vera C. Rubin Observatory, the Dark Energy Spectroscopic Instrument (DESI) on the Mayall telescope in Arizona, and the Gemini North and South and SOAR telescopes, each contributing a different piece: imaging, spectroscopy, and repeated follow-up as the explosion evolved. Xander Hall, a graduate student on the Carnegie Mellon team, credited DESI’s ability to be redirected quickly: “DESI’s spare-fiber program gave us the opportunity to return to SN 2026gzf repeatedly and follow how its spectrum changed as the explosion evolved.”

That kind of coordinated, fast-turnaround follow-up is itself part of the story. Shock breakouts last hours at most, which means catching one at all depends on an X-ray satellite flagging it within minutes and ground-based telescopes being available to look immediately afterward. Gokul Srinivasaragavan, a recent University of Maryland PhD graduate on the team, framed the case as a first of its kind rather than a one-off: “This is the first time we’ve mapped out the pre-explosion environment of a star that has been stripped of hydrogen and helium. Going forward, I’m excited to observe more shock breakout events in similar detail to test if all stripped stars have a similar ‘lifestyle’ prior to collapse.”

One explosion, an open question

It’s worth being precise about what one well-observed supernova can and can’t settle. SN 2026gzf shows that a broad-lined Type Ic supernova can occur without a detected gamma-ray burst or jet, and that a choked jet is a physically consistent explanation for what was and wasn’t seen. It does not establish how common choked jets are among this class of supernova generally, since the sample of confidently caught shock breakouts now stands at two. Whether SN 2008D and SN 2026gzf sit at opposite ends of a spectrum of outcomes, or whether most events of this kind look more like one than the other, isn’t something a single additional data point can answer.

What it does change is the assumption that finding no jet means finding no jet was ever there. For SN 2026gzf, the evidence points instead to a jet that started, and lost the fight to get out.