How Did the Longest Gamma‑Ray Burst Ever Rewrite the Rules?

Welcome, dear readers of FreeAstroScience.

What kind of cosmic engine can stay “on” for days, hurling out gamma rays with the power of a thousand Suns shining for 10 billion years? That’s the riddle behind GRB 250702B, the longest gamma‑ray burst ever recorded. This article, written by FreeAstroScience only for you, walks through what we know, what still puzzles us, and why this one event is keeping astronomers awake at night. Stay with us to the end: the story gets stranger the deeper we go.

What are gamma‑ray bursts, and where does this one fit?

Gamma‑ray bursts (GRBs) are violent flashes of high‑energy photons, usually lasting from a fraction of a second to a few minutes. They’re thought to come from:

Short GRBs (≲2 s): mergers of neutron stars or neutron‑star–black‑hole binaries.
Long GRBs (≳2 s): collapse of massive stars (“collapsars”) into stellar‑mass black holes, launching relativistic jets.
Ultralong GRBs: rare events with prompt emission lasting thousands of seconds, possibly from stars with very large radii, like blue supergiants.

In most cases, we see:

A brief, erratic gamma‑ray flash, then
A smoother, fading afterglow in X‑rays, optical, and radio, powered by a decelerating relativistic jet ploughing into surrounding gas.

Astronomers have catalogued roughly 15,000 GRBs so far, about one per day, but GRB 250702B stands out dramatically.

Why is GRB 250702B so extreme?

How long did it actually last?

On 2025 July 2, the Fermi Gamma‑ray Burst Monitor (GBM) saw multiple high‑energy triggers—labelled GRB 250702D, B, C, and E—all from the same patch of sky.

Then the surprises started:

Konus‑Wind found overlapping high‑energy emission episodes stretching over >25,000 seconds (about 7 hours) in gamma rays alone. ,[2]]
The Einstein Probe (EP) detected soft X‑ray emission almost 24 hours before the first Fermi trigger, on July 1.
Swift’s BAT, in survey mode, also caught lingering hard X‑ray/gamma emission around and after the main triggers.

So depending on how you count:

Prompt high‑energy flaring: >25 ks (≈7 hours) in gamma rays. ,[2]]
Central engine activity, traced by soft X‑rays: ≳3 days in the observer frame (≳1.5 days in the rest frame).

That’s far beyond classical GRBs and even longer than the previous “record‑holder”, GRB 111209A, with ~15 ks of gamma‑ray emission and ~25 ks engine activity. ,[2]]

How bright and energetic was it?

A team using JWST spectra measured the host‑galaxy redshift as z = 1.036, putting GRB 250702B about 8 billion light‑years away—light left the source long before the Sun existed. ,[3]]

Konus‑Wind data give an isotropic‑equivalent gamma‑ray energy:

(E_{\gamma,\mathrm{iso}} \approx 1.4 \times 10^{54},\mathrm{erg}) (1–10,000 keV band).

That’s:

At the upper end of long‑GRB energetics.
Comparable to the brightest known bursts, like GRB 221009A.

Even more striking, one analysis estimated the outburst released the equivalent energy of ~1000 Suns shining for 10 billion years in high‑energy radiation.

So we have a burst that is:

Very long,
Very energetic, and
Coming from cosmological distance.

That already hints we’re dealing with a powerful relativistic jet, not some local oddball.

How did astronomers follow a burst that outlived their telescopes?

No single instrument could track GRB 250702B from start to finish. Astronomers stitched together a timeline from a fleet of spacecraft and ground‑based telescopes: ,[2],

Einstein Probe (EP) – soft X‑ray monitoring before any gamma‑ray triggers, revealing early activity nearly one day before.
Fermi GBM & Konus‑Wind – multiple, overlapping high‑energy triggers giving the 7‑hour prompt duration and MeV photons. ,[2]]
Swift/BAT & XRT – hard X‑ray survey detections plus a fading X‑ray afterglow from ~0.5 to ~45 days.
NuSTAR – sensitive hard‑X‑ray follow‑up (3–79 keV) at ~1.6, 5.7, and 10 days.
Chandra – deep X‑ray imaging at ~38 and 66 days, confirming the source was still there and fading.
VLT, Keck, Gemini, Magellan, HST – near‑infrared (NIR) and optical imaging of the obscured afterglow and the host galaxy.
Radio arrays (VLA, ALMA, etc.) – radio afterglow detections sampling the jet’s interaction with its environment.

Eric Burns summed it up sharply: no high‑energy monitor in space was designed to follow something this long; only the combined power of multiple spacecraft and telescopes made sense of the event.

That patchwork view is part of the fun: we’re watching a jet evolve across the entire spectrum, for weeks, in real time.

What did the X‑rays say about the hidden engine?

A strange “prelude” in soft X‑rays

The Einstein Probe reported soft X‑ray emission (0.5–4 keV) about 24 hours before Fermi’s first gamma‑ray trigger.

That early emission:

Rose smoothly toward the main activity,
Doesn’t look like a brief “precursor” spike,
And simply doesn’t fit standard long‑GRB models, where the prompt phase is tied closely to core collapse.

It tells us the engine was already active well before the violent gamma‑ray flares.

A stubborn engine that refused to shut off

From 0.5 to 65 days after the first Fermi trigger, Swift/XRT, NuSTAR, and Chandra saw a fading X‑ray source. The flux roughly followed a power‑law decay:

\(F_X(t) \propto t^{-\alpha}\), with \(\alpha \approx 1.7–1.9\) depending on the exact start time adopted.

On top of that smooth decline, two things stood out:

Short‑timescale flares in X‑rays at ≲2 days:
- NuSTAR saw variability on ~1–2 ks timescales, with fractional variability ΔT/T < 0.03.
- Swift/XRT saw even faster flares (~150 s) at ~0.54 days.
Continued hard‑X‑ray detections (Swift/BAT survey) hours after the last GBM trigger, hinting at late internal activity.

Those fast fluctuations are too rapid to arise from the external shock alone. They demand a still‑active central engine—almost certainly an accreting black hole—feeding the jet for at least 3 days in the source frame.

That’s a key aha moment: we’re not just watching debris cooling down; we’re watching an engine that keeps turning on fresh jets or shells long after any usual GRB would have gone quiet.

“GRB‑like” X‑ray brightness, not “TDE‑like”

When you compare the X‑ray luminosity at 11 hours (rest frame) to the gamma‑ray energy, GRB 250702B falls right on the usual GRB correlation:

More energetic prompt emission → brighter 11‑hour X‑ray afterglow.

Relativistic tidal disruption events (TDEs) around supermassive black holes usually sit in a different part of that diagram: they have very luminous, long‑lived X‑rays without the same “GRB‑style” prompt correlations.

In contrast, this event looks statistically like a GRB in its X‑ray–to–gamma relation—even though its engine lives much longer than typical collapsars.

What kind of medium did the jet blast through?

To understand the environment, teams fitted the broadband afterglow—radio, NIR, and X‑ray—with standard forward‑shock (and sometimes reverse‑shock) models.

The picture that emerges is:

The external density falls off like a stellar wind, approximately
(\rho(r) \propto r^{-2}).
The afterglow requires a very energetic, ultrarelativistic jet:
- Isotropic‑equivalent kinetic energy
  (E_{\mathrm{kin,iso}} \sim 10^{54-55},\mathrm{erg}).
- Initial Lorentz factor (\Gamma_0 \gtrsim 100), and possibly a few hundred if you insist the jet does not spread early.
The jet opening angle is tiny, likely ≲1°, implying strong collimation.

Here’s a compact summary based on current modeling: ,[3]]

**Table 1 – Inferred jet and environment properties of GRB 250702B** (observer‑frame quantities converted using z = 1.036)
Quantity	Symbol	Approximate value	Comment
Redshift	z	1.036	JWST host spectroscopy ,[3]]
Isotropic gamma‑ray energy	E_γ,iso	≈ 1.4 × 10⁵⁴ erg	Konus‑Wind + Fermi GBM
Isotropic kinetic energy	E_kin,iso	∼ 10^54–55 erg	Afterglow fits ,[3]]
Beaming‑corrected total (γ + kinetic)	E_true	≈ few × 10⁵⁰ erg	Assuming very narrow jet
Initial Lorentz factor	Γ₀	≳ 100 (possibly > 300)	Needed to decelerate early and avoid opacity
Jet half‑opening angle	θ_j	≲ 0.02 rad (~1°)	Jet break must occur before observations or be very late
External density profile	k in ρ ∝ r^−k	k ≈ 2	Wind‑like environment
Line‑of‑sight visual extinction (afterglow)	A_V,GRB	≈ 5–9 mag	Heavily obscured afterglow

A wind‑like environment is exactly what you expect around a massive star shedding mass, not around a lone intermediate‑mass black hole sitting in the outskirts of a galaxy. That one detail already leans us toward a stellar‑mass progenitor.

What does the host galaxy look like?

Here the story gets even richer—and messier. High‑resolution NIR imaging with HST, VLT, Keck, Gemini, and Magellan reveals a very odd host: ,[3]]

Massive and dusty:
- Stellar mass (M_* \approx 10^{10.66},M_\odot).
- Host‑integrated extinction (A_{V,\text{host}} \approx 0.9) mag.
Morphology:
- Strongly asymmetric, with concentration–asymmetry (CAS) parameters suggesting a major merger.
- The NIR light can be modeled either as:
  - Two Sérsic components, like two galaxies in the process of merging, or
  - A single galaxy bisected by a dust lane. ,[3]]
Burst location:
- Projected offset ≈0.7″, or ≈5.8 kpc at z ≈ 1.
- Clearly off‑nuclear, not sitting on a central supermassive black hole. ,[3]]

In the CAS diagram, this host’s asymmetry (A ≈ 0.9) is far higher than typical GRB hosts and also higher than the host of the canonical jetted TDE Sw J1644+57.

So we likely have:

A massive, merging, dusty system, and
A burst happening away from the galaxy center, but still inside a dense, complex environment.

Whatever happened, it did not occur at the tidy heart of a quiet elliptical hosting a supermassive black hole.

How dusty is the line of sight—and why does that matter?

After correcting for Milky Way dust, both X‑ray and NIR data demand heavy intrinsic extinction:

From broadband SED fitting, the afterglow requires
(A_{V,\text{GRB}} \sim 5–9,\mathrm{mag}). ,[3]]
Host SED fitting gives a much smaller global (A_{V,\text{host}} \sim 0.9,\mathrm{mag}).

That mismatch tells us that extra dust lies either:

Very close to the progenitor (circumstellar), or
In a dense dust lane along our line of sight.

X‑ray absorption provides the hydrogen column (N_H). Combining X‑ray and NIR fits, one can estimate a gas‑to‑dust ratio:

\( \frac{N_{H,z}}{A_{V,z}} \approx (4.5 \pm 1.0) \times 10^{21}\,\mathrm{cm^{-2}\,mag^{-1}} \),

which is roughly twice the Milky Way value and similar to what’s seen in the dusty GRB 130925A and the jetted TDE Sw J1644+57.

So GRB 250702B belongs to that “hidden” population of heavily obscured relativistic transients that you will simply miss without deep NIR follow‑up and good X‑ray coverage.

What could have caused such a bizarre event?

Here’s where the arguments start, and where we should be honest: there is no single model that cleanly explains all the data yet. You can feel the field thinking out loud in the literature.

Let’s walk through the leading options.

Option 1: A very unusual collapsar (a dying massive star)

This is the “extended GRB” idea: maybe GRB 250702B is an ultralong GRB from a massive star with a big envelope (like a blue supergiant) or from a helium‑star merger. ,[2],

Why it’s attractive:

The host looks star‑forming and dusty; the burst occurs off‑nucleus.
The external medium is wind‑like (ρ ∝ r⁻²), as expected around massive stars.
The jet’s Lorentz factor, opening angle, and beaming‑corrected energy all look typical of GRBs when corrected for collimation. ,
The event obeys GRB prompt–afterglow correlations like (E_{\gamma,\text{iso}}–L_{X,11}).

Why it’s problematic:

Standard collapsar models struggle to power day‑scale engine activity; they usually shut off within minutes to at most a few hours.
The 24‑hour‑earlier soft X‑ray emission is very hard to reconcile with a single, smooth core‑collapse jet breakout.

One could imagine a very extended envelope plus a cocoon or repeated fallback episodes, but that starts to feel like piling tweaks on a model that wasn’t designed for this timescale. The helium‑star merger subtype—where a black hole merges with a helium core inside a binary—is one way to stretch the engine, but even then, the physics has to work hard.

Option 2: A tidal disruption event by a massive black hole

Tidal disruption events (TDEs) happen when a star strays too near a massive black hole and gets shredded. In a jetted TDE, some of the fallback gas powers a relativistic jet aimed at us, producing bright, long‑lasting X‑rays and sometimes gamma rays. Classic examples include Sw J1644+57 and AT 2022cmc.

Could GRB 250702B be a relativistic TDE?

Clues in favor:

The multi‑day engine activity and central‑engine–dominated X‑ray variability resemble jetted TDEs.
The heavy obscuration and high gas‑to‑dust ratio are similar to Sw J1644+57.

Serious obstacles:

Location problem:
- The transient is offset by ~5.8 kpc from the galaxy center.
- That rules out a standard supermassive black hole (SMBH) at the nucleus, unless we’re dealing with an exotic “wandering SMBH” scenario.
Timescale problem for a main‑sequence star:
A useful scaling for the fallback time of the most bound debris is:

\( t_{\rm peak} \approx 25\,\mathrm{d}\, \left(\frac{M_\bullet}{10^6\,M_\odot}\right)^{1/2} \),

where (M_\bullet) is the black‑hole mass and we assume a solar‑type star.

To get (t_{\rm peak} \lesssim 1.5) days (source frame), you need
(M_\bullet \lesssim 10^3,M_\odot): an intermediate‑mass black hole (IMBH).

That’s possible—especially in a merging, messy galaxy—but it’s already a niche scenario.
Variability timescale vs. black‑hole mass:
The shortest variability time in the gamma rays is ≲1 s.
For a massive BH, the light‑crossing time sets a rough lower bound on variability, (t_s \sim 2GM/c^3). For (M \sim 10^5–10^6,M_\odot), that’s much longer than a millisecond engine. Matching the observed ≲1 s minimum variability pushes you toward either:
- A relatively small IMBH (≲ few × 10⁴ M⊙), or
- A “jets‑in‑a‑jet” magnetic reconnection picture, like in blazars, where sub‑structures vary faster than the global light‑crossing time.
X‑ray luminosity and evolution:
The X‑ray light curve of GRB 250702B looks more like a GRB afterglow than like known jetted TDEs, which stay at very high X‑ray luminosity for months and show dramatic, deep dips.

A white‑dwarf TDE by an IMBH was also tested and found wanting, because the fallback timescales from simulations are too short—hours, not days—to match the observed long activity, unless you add extra complications like repeated partial disruptions on tight orbits.

So a main‑sequence–star TDE by an IMBH remains technically possible, but it feels contrived compared to GRB‑like options.

Option 3: A micro‑TDE – a star shredded by a stellar‑mass black hole

Here we change scale completely: instead of a massive or intermediate‑mass BH, imagine a stellar‑mass black hole (say, 3–10 M⊙) tearing apart a companion star. This is sometimes called a micro‑TDE.

In such a system:

The orbital separation is small, so encounters can be deep.
The fallback timescale and accretion rate can naturally span days, not minutes.
The engine is still a stellar‑mass BH, so second‑scale variability is easy to produce.

Beniamini and collaborators argue that a micro‑TDE can provide a hybrid central engine: TDE‑like in terms of multi‑day accretion, but GRB‑like in terms of Lorentz factor, jet structure, and environment.

This scenario:

Matches the long engine duration and X‑ray flares.
Respects the GRB‑style Lorentz factor and energetics.
Fits naturally into a wind‑like environment, because the disrupted star could have been a compact, stripped star in a binary.

We don’t yet have a smoking gun feature that uniquely screams “micro‑TDE”, but in terms of overall consistency, this is one of the front‑runner ideas.

Option 4: A black‑hole–helium‑star merger

Another proposed hybrid involves a stellar‑mass black hole spiralling into a helium star companion—a system originally invoked to explain some ultralong GRBs.

Picture this:

The hydrogen envelope has already been stripped, leaving a compact helium star and a close black‑hole companion.
The BH plunges into the helium star, accretes furiously from inside its envelope, and powers a long‑lived jet.
The interaction may trigger an unusual, possibly faint supernova.

For GRB 250702B:

The BH + He‑star merger can, in principle, sustain activity for ∼day timescales.
Any associated supernova could be hidden by the thick dust, which is consistent with the lack of a clear optical supernova bump. ,[3]]

Again, we lack direct proof, but this scenario is flexible enough to explain:

The long engine duration,
The massive, star‑forming, merging host, and
The off‑center, dusty location. ,[3]]

So what does GRB 250702B really teach us?

In some sense, GRB 250702B is less about putting a badge—“GRB”, “TDE”, “micro‑TDE”—on one object and more about what it says about our categories.

A few lessons stand out:

Astrophysical engines don’t care about our boxes.
We like to separate “long GRBs from collapsars” and “jetted TDEs from black‑hole disruptions” as if they were disjoint families. GRB 250702B sits near the border and whispers, “maybe nature blends these mechanisms.”
Soft X‑ray monitoring changes the game.
Without Einstein Probe’s early soft‑X‑ray data, we’d have completely missed that the engine started a full day before the first gamma‑ray trigger. Future wide‑field soft‑X‑ray missions will almost certainly reveal more events with long, quiet “ramps” before the fireworks.
Heavily obscured relativistic events are commoner than we see.
With (A_V \sim 5–9) mag, this afterglow would have been invisible in standard optical follow‑up. Only fast, deep NIR observations plus X‑rays revealed it. That means our current GRB and TDE samples are biased toward “clean‑line‑of‑sight” events. ,[3]]
Host galaxies matter.
The combination of a merging, massive, dusty galaxy and an off‑nuclear, wind‑like environment points strongly toward stellar‑mass black holes living in messy star‑forming regions, not just at galaxy centers.
There is real uncertainty—and that’s healthy.
Teams led by O’Connor and Carney explicitly conclude that GRB 250702B “does not neatly fit” known classes and may arise from a novel progenitor. ,[3]]

That honesty is part of what makes this event scientifically valuable: it forces theory to stretch, not just to repaint an old model.

What should we, as curious humans, take away from all this?

If we step back from the details, GRB 250702B is a reminder that:

The Universe is more inventive than our current textbooks.
Our classifications—short, long, ultralong, TDE—are stepping stones, not final answers.
Progress comes from letting data challenge comfortable stories.

There’s a poetic twist here. Francisco Goya once etched the phrase “The sleep of reason breeds monsters.” In his context, it was a warning about superstition and political darkness. In our context, there’s a nice double meaning:

If we let reason fall asleep, we slide into ignorance and fear.
But when we keep reason awake, we can actually study the monsters—the black holes, jets, and cosmic disruptions—that would otherwise just terrify us.

GRB 250702B is one of those monsters. Not in the horror‑movie sense, but in the sense of an event so large and strange that it forces reason to stay sharp. It stretches our models of black‑hole accretion, stellar death, and relativistic jets, and it shows why we need every photon—from soft X‑ray to radio—to make sense of the sky.

We’ll almost certainly see more events like this as Einstein Probe, JWST, next‑generation radio arrays, and future gamma‑ray missions keep watching. Each one will either fill in this picture or rip up parts of it. Either way, curiosity wins.

This post was written for you by FreeAstroScience.com, which specializes in explaining complex science in simple, honest language. Our aim is to inspire your curiosity about the cosmos—and to keep reason awake, because the sleep of reason breeds monsters.

Come back soon; the Universe has more strange stories waiting.

What Does the Longest Gamma‑Ray Burst Ever Reveal About Black Holes?

How Did the Longest Gamma‑Ray Burst Ever Rewrite the Rules?

What are gamma‑ray bursts, and where does this one fit?