Hubble Tension Explained

 Hubble Tension — Is Our Model of the Universe Wrong?

We can measure the Universe from nearby stars and from light released more than 13 billion years ago. Both methods are extraordinarily precise. They still disagree on how fast the cosmos is expanding today.

Two precise views of the same Universe point to different expansion rates. Somewhere between the early cosmos and the galaxies around us, something does not quite add up.

The Universe Has a Number That Refuses to Behave

For much of modern cosmology, the expansion rate of the Universe was frustratingly uncertain. Astronomers agreed that space was expanding; they disagreed sharply about how fast. As telescopes improved and samples grew, the error bars finally began to collapse.

That should have settled the argument. Instead, precision revealed a new one.

One family of measurements now places the present expansion rate — the Hubble constant, H₀ — near 73 kilometers per second per megaparsec. Another, anchored in the early Universe and interpreted through the standard ΛCDM model, lands near 67–68. The gap is only about 8 percent. In a field that now measures key quantities at roughly percent-level precision, it is enormous.

This is the Hubble tension. And it has become harder to dismiss. In 2025, ACT DR6 combined with DESI DR2 gave H₀ = 68.43 ± 0.27 km/s/Mpc within ΛCDM. An expanded SH0ES analysis using JWST and HST, published the same year, found 73.49 ± 0.93 from Cepheids, or 73.18 ± 0.88 when Cepheid and TRGB calibrations were combined. These are not relics of an old Planck-versus-Hubble dispute. They are modern datasets, with very different systematics, still pulling the cosmic scale apart.

Yet even that description is too simple. Another JWST program, the Chicago–Carnegie Hubble Program, obtained a combined value near 70 with larger systematic uncertainty. Strong gravitational lenses prefer a central value above 70 but are not yet precise enough to decide the dispute. Megamasers give an independent geometric result near 74, again with larger uncertainty. Standard sirens from gravitational waves are still too broad to choose a side.

So the mystery is no longer “Planck versus Hubble.” It is a detective story with many witnesses: pulsating stars, exploding white dwarfs, primordial sound waves, orbiting water molecules, warped quasar light and colliding black holes. Somewhere in that chain, either a measurement is misleading us or the model connecting the early and late Universe is incomplete.

What Does H₀ Actually Mean?

The Hubble constant is the present value of the cosmic expansion rate. In the nearby Universe, where a simple approximation works, a galaxy’s recession velocity is proportional to its distance: v ≈ H₀d.

The strange unit — kilometers per second per megaparsec — becomes intuitive with an example. A megaparsec is about 3.26 million light-years. If H₀ were exactly 70 km/s/Mpc, then two points separated by one megaparsec would, on average, be carried apart by cosmic expansion at about 70 km/s; at 100 megaparsecs the corresponding Hubble-flow speed would be about 7,000 km/s.

This does not mean galaxies are simply flying through static space from one central explosion. On large scales, the metric itself evolves: distances between unbound regions grow as the Universe expands. Nearby galaxies can still move toward us because local gravity overwhelms the Hubble flow — Andromeda is the familiar example.

Another subtlety matters enormously for this debate: H₀ is not the expansion rate at every epoch. The expansion rate H(z) changes with time as radiation, matter and dark energy contribute differently. H₀ is H(z) today. Therefore a measurement made from ancient light can constrain H₀ only after a model tells us how to connect the early Universe to the present.

One Constant, Many Ways to Measure It

A 2024 Annual Review captured the modern problem in its title: “A Tale of Many H₀.” Cosmologists can approach the expansion scale through Cepheids, red giant stars, JAGB stars, Type Ia supernovae, the cosmic microwave background, baryon acoustic oscillations, gravitational lenses, megamasers, stellar ages and gravitational-wave standard sirens. These methods do not share the same weaknesses. That is precisely why comparing them is so powerful.

If the measurements are unbiased and the cosmological model is correct, all of those roads should eventually meet. They do not. Several precise determinations cluster around 67–69; the best-known local Cepheid ladder remains near 73; less precise techniques stretch across the gap. The useful question, then, is not “Which number should we believe?” but “What does each method actually measure, and what assumptions turn that observation into H₀?”

Method One: Build a Ladder from Our Galaxy to the Hubble Flow

The classic local strategy is called the cosmic distance ladder. No single object can provide precise distances from our immediate neighborhood to hundreds of millions of light-years, so astronomers build a chain in which one rung calibrates the next.

The first rung uses geometry. Parallax from missions such as Gaia measures the apparent shift of nearby stars as Earth changes position around the Sun. Eclipsing binary stars in the Large Magellanic Cloud provide another geometric anchor. The orbiting water megamasers in NGC 4258 supply an extraordinarily valuable galaxy-scale geometric distance.

Those anchors calibrate Cepheid variable stars. Cepheids pulsate with a period related to their intrinsic luminosity. Measure the period, infer how bright the star really is, compare that with how bright it appears, and distance follows. The relation is powerful enough to carry the ladder into galaxies tens of millions of light-years away.

The next rung is the Type Ia supernova. These stellar explosions are not perfectly identical candles, but their luminosities can be standardized using the shape and color of their light curves. A nearby galaxy containing both Cepheids and a well-observed Type Ia supernova tells astronomers the supernova’s absolute luminosity. Then Type Ia supernovae much farther away can map the smooth Hubble flow, where local peculiar velocities are a smaller fraction of the total redshift.

The SH0ES collaboration’s major 2022 analysis used Cepheids in hosts of 42 Type Ia supernovae and obtained H₀ = 73.04 ± 1.04 km/s/Mpc. The team tested dozens of analysis variants involving anchors, metallicity, dust, supernova samples, redshift cuts and other choices. The high value persisted.

Astronomers cannot measure the Universe with a single ruler. Each rung of the cosmic distance ladder calibrates the next, eventually allowing Type Ia supernovae to carry the measurement deep into the expanding Universe.

JWST Was Supposed to Expose the Hidden Error

One of the most plausible concerns about the Cepheid ladder was stellar crowding. At the distances of supernova-host galaxies, Hubble Space Telescope images can blend a Cepheid with neighboring unresolved stars. Extra light could make the Cepheid appear brighter and therefore closer than it really is. If distances were underestimated systematically with distance, H₀ could be biased high.

JWST was built with a much larger mirror and superb infrared resolution. It can separate stellar backgrounds that Hubble cannot. This turned Webb into an unusually clean experiment: repeat crucial Cepheid measurements with a different telescope and ask whether Hubble’s crowding corrections were hiding a large bias.

The SH0ES results have not found the required bias. Their JWST work found substantially reduced scatter but no distance-dependent offset capable of lowering H₀ from about 73 to about 68. The 2025 “Perfect Host” study pushed the test further using NGC 3447, where Cepheids could be compared across ordinary crowded regions and a young companion with almost no old-star background. Again, the measurements did not reveal the missing crowding error. Across 19 JWST-observed SH0ES hosts, the team reported no evidence for a photometric bias relative to HST large enough to resolve the tension.

That removes one of the cleanest proposed explanations. It does not prove new physics. Webb tests particular observational weaknesses; it cannot certify every rung of the ladder, every supernova calibration or every assumption connecting them.

Then JWST Complicated the Story

The Chicago–Carnegie Hubble Program deliberately approached the local scale using three stellar indicators observed with JWST: Cepheids, the tip of the red giant branch (TRGB), and J-region asymptotic giant branch stars (JAGB). All three were tied to the same geometric anchor, NGC 4258, and then connected to Type Ia supernova hosts.

Their 2024 status report found H₀ = 72.05 ± 1.86(stat) ± 3.10(sys) from Cepheids, 69.85 ± 1.75 ± 1.54 from TRGB, and 67.96 ± 1.85 ± 1.90 from JAGB. Combining the methods gave 69.96 ± 1.05(stat) ± 1.12(sys) km/s/Mpc.

This is exactly the kind of result that makes the tension scientifically valuable. There is no single “JWST value” of H₀. Webb improves the observations, but calibrator choice, host selection and analysis still matter. In the CCHP sample, TRGB and JAGB distances agreed at roughly the one-percent level while their central H₀ values sat below the Cepheid result. The telescope reduced one source of uncertainty and exposed the importance of others.

The local side of the Hubble tension therefore contains a tension of its own: some highly precise ladder analyses remain near 73, while other carefully constructed JWST routes land closer to the middle.

Method Two: Use the Oldest Light in the Universe

The lower value begins with an entirely different experiment. The cosmic microwave background is radiation released when the Universe was about 380,000 years old, after the primordial plasma cooled enough for electrons and nuclei to form neutral atoms and photons could travel freely.

Before that moment, photons and ordinary matter behaved like a coupled fluid. Gravity tried to compress overdense regions while radiation pressure pushed back, creating acoustic oscillations. The CMB preserves a frozen pattern of those oscillations. Its temperature and polarization power spectra contain a sequence of acoustic peaks whose positions and heights encode the baryon density, dark-matter density, primordial fluctuations, geometry and other cosmological information.

Planck measured this pattern with extraordinary precision. Under the six-parameter flat ΛCDM model, its final analysis inferred H₀ ≈ 67.4 ± 0.5 km/s/Mpc. Crucially, this is not a direct local measurement of today’s recession velocities. It is a model-based inference: fit the early-Universe pattern, determine the parameters, and evolve the model forward to today.

That distinction is not a weakness unique to Planck; it is the central clue. If ΛCDM is incomplete, the CMB can be measured perfectly while the inferred H₀ is wrong.

Planck Is No Longer Alone: ACT DR6 Strengthens the Low-H₀ Side

For years, skeptics could reasonably ask whether the tension somehow reflected Planck-specific data or analysis. The Atacama Cosmology Telescope has made that explanation harder.

ACT DR6 measured temperature and polarization anisotropies across a huge fraction of the sky with exceptionally strong small-scale polarization sensitivity. Its 2025 ΛCDM analysis found that the CMB spectra are well described by the standard model. When ACT and large-scale Planck information were combined with CMB lensing and DESI DR1 BAO, the result was H₀ = 68.22 ± 0.36 km/s/Mpc. Replacing DR1 with DESI DR2 tightened it to 68.43 ± 0.27.

The number is slightly higher than the classic Planck value, but nowhere near 73. More importantly, a modern CMB experiment with different instrumentation still points to the low-H₀ region when interpreted through ΛCDM. The early-Universe side is therefore not resting on one satellite.

The heart of the Hubble tension: early-Universe observations interpreted through ΛCDM favor H₀ near 67–68 km/s/Mpc, while the SH0ES distance ladder remains near 73 km/s/Mpc.

The Hidden Player: The Sound Horizon

To understand why new physics is so difficult, we need one more concept: the sound horizon. Before recombination, acoustic waves traveled through the photon–baryon plasma. The maximum distance those waves could travel before the plasma released the CMB became a physical standard ruler.

The same primordial acoustic scale later appears in the distribution of galaxies as baryon acoustic oscillations, or BAO. Surveys such as DESI measure the apparent size of that ruler at different redshifts, allowing cosmologists to reconstruct the relative expansion history with remarkable precision.

This reveals a useful way to think about the tension. The shape of the late-time expansion history is broadly well measured. The fight is largely about the absolute calibration of the cosmic ruler and candle system. To raise the CMB/BAO-inferred H₀ substantially, many proposed models must make the early-Universe sound horizon smaller. But shrinking that ruler without spoiling the CMB peaks, primordial nucleosynthesis and structure formation is extraordinarily difficult.

DESI: A New Map of Expansion — but Not a Simple Solution

DESI measures millions of galaxies and quasars to trace BAO across cosmic time. Its first three years produced the most precise BAO measurements yet. When combined with CMB and supernova data, DESI DR2 strengthened evidence in some dataset combinations for an evolving dark-energy equation of state rather than a perfectly constant cosmological constant.

That sounds like exactly the kind of crack in ΛCDM that the Hubble tension needs. But cosmology is rarely that cooperative. Dynamical dark energy at late times does not automatically lift H₀ to the SH0ES value. Some analyses find that the DESI-preferred evolution can actually coexist with a relatively low H₀, leaving or even sharpening the original disagreement.

The story also changed again in July 2026. DESI released a much more precise full-shape analysis of the Lyman-alpha forest. The new central value moved toward the standard ΛCDM prediction compared with the earlier BAO-only Lyman-alpha result. DESI itself emphasized that this could mean the evolving-dark-energy hints weaken with more information — or that a more complicated model is required.

The editorial lesson is simple: “DESI discovered changing dark energy” goes beyond the evidence. DESI has found intriguing, dataset-dependent departures from a cosmological constant; newer probes are testing whether that pattern survives as the analysis becomes more complete.

How Significant Is the Tension?

The answer depends on exactly which measurements are compared. The familiar headline compares the high-precision SH0ES ladder with the ΛCDM prediction from early-Universe datasets, producing a discrepancy around five sigma in earlier work. The expanded 2025 SH0ES JWST+HST analysis reported a combined Cepheid+TRGB result about six sigma above the ΛCDM+CMB expectation.

But sigma is not a magic detector of new physics. A five-sigma random fluctuation is extraordinarily unlikely under an ideal statistical model, yet systematic errors are not guaranteed to behave like random Gaussian noise. A shared calibration bias can survive enormous sample sizes. Model assumptions can also produce apparently tiny statistical errors while leaving a deeper theoretical uncertainty untouched.

That is why sigma cannot settle the story by itself. The decisive test is whether measurements with genuinely different failure modes begin to converge.

First Independent Witness: Water Megamasers

Some galaxies contain thin disks of water-bearing molecular gas orbiting supermassive black holes. Microwave maser emission from these disks allows astronomers to map orbital velocities and accelerations and derive geometric distances without Cepheids or the CMB.

The Megamaser Cosmology Project combined six maser galaxies and obtained H₀ = 73.9 ± 3.0 km/s/Mpc. Its central value is strikingly close to the high local ladder, but the uncertainty is several times larger than SH0ES. It therefore supports the possibility of a high H₀ without yet delivering a decisive verdict.

Second Independent Witness: Gravitational Lenses

General relativity gives another cosmic clock. A massive foreground galaxy can bend light from a distant variable quasar into multiple images. Because each image follows a different path through a different gravitational potential, brightness variations arrive at different times. Those time delays depend on combinations of cosmological distances and are sensitive to H₀.

The difficulty is the mass model of the lens. Different distributions of visible and dark matter can reproduce similar lensing images while changing the inferred time-delay distance — the famous mass-sheet degeneracy.

TDCOSMO 2025 used eight strongly lensed quasars, improved stellar kinematics from JWST, Keck and the VLT, and deliberately conservative treatment of lens mass profiles. Combined with Pantheon+ matter-density information, it found H₀ = 72.1 +4.0/−3.7 km/s/Mpc in flat ΛCDM. The central value is high, but the roughly five-percent precision comfortably overlaps both camps. That makes lensing an independent referee still waiting for a larger sample.

Third Independent Witness: Gravitational-Wave Sirens

Binary neutron-star and black-hole mergers create gravitational waves whose waveform contains an absolute luminosity-distance scale. In principle, nature provides the ruler directly through general relativity. If the redshift of the host galaxy can also be determined — either from an electromagnetic counterpart or statistically from galaxy catalogs — the event becomes a “standard siren” for H₀.

The attraction is obvious: no Cepheid period–luminosity relation, no supernova absolute magnitude and no CMB sound horizon. The present limitation is statistics and source identification. A 2026 analysis combining 142 GWTC-4 standard-siren events with TDCOSMO lensing remained broad enough to be consistent with both Planck-like and SH0ES-like values. Future observing runs could transform this method because uncertainty should shrink as catalogs grow.

If both sides of the Hubble tension are substantially correct, the missing ingredient may lie somewhere in the physics connecting the young Universe to the cosmos we observe today.

Could We Simply Live in a Cosmic Void?

One intuitive proposal is that the Milky Way lies inside an unusually underdense region. Matter outside would gravitationally pull more strongly than matter inside, making nearby galaxies appear to recede faster than the global average. A local “Hubble bubble” could therefore bias a local H₀ upward.

Cosmic variance certainly affects local expansion measurements at some level, and the nearby Universe is not perfectly homogeneous. But the void required to explain the full 67-versus-73 gap would be unusually large and deep. Modern analyses and reviews generally find that realistic local structure can contribute only a small fraction of the tension. The 2026 decade review describes a simple local void solution as effectively ruled out as the dominant explanation.

Suppose the Measurements Are Right. What Has to Change?

A successful new theory has an unforgiving job. It must raise the inferred H₀ without ruining the CMB acoustic peaks, BAO distances, primordial light-element abundances, galaxy clustering, weak lensing, supernova distances and the many other successes of ΛCDM. Moving one number is easy. Moving it while leaving the rest of cosmology standing is the real test.

Most proposals attack one of two places. Early-time solutions change the physics before or around recombination, altering the primordial ruler used to calibrate the Universe. Late-time solutions change the recent expansion history or the way supernova distances are calibrated. More radical ideas modify gravity or allow new interactions in the dark sector.

Early Dark Energy: Shrink the Primordial Ruler

Early dark energy is designed to intervene briefly before recombination. For a short period it contributes extra energy density, speeds up the expansion and reduces the sound horizon — the maximum distance primordial pressure waves could travel before the CMB formed. If that physical ruler is smaller than ΛCDM assumes, the same observed angular pattern can be reconciled with a larger H₀ today.

That is why the idea attracted so much attention: it targets the calibration point where the early-Universe inference is set, rather than forcing a large distortion into the well-measured late-time expansion curve.

But the extra component changes more than one ruler. It affects the CMB peak structure and the growth of matter fluctuations. Some datasets constrain it strongly. ACT DR6’s 2025 extended-model analysis found no statistically significant preference for departures from baseline ΛCDM; in several extensions designed to raise H₀, inferred values remained around 69–70 rather than 73.

There is, however, an active methodological debate. A separate 2025 analysis of ACT DR6 and DESI DR2 argued that early dark energy remains viable and that profile-likelihood methods can permit H₀ around 71 without SH0ES, with the residual SH0ES discrepancy reduced relative to standard ΛCDM. The important point is not that one paper has “won,” but that EDE is now testable against increasingly constraining data rather than functioning as an unconstrained escape hatch.

Other Early-Universe Possibilities

Additional relativistic particles — often summarized through the effective number of relativistic species, N_eff — can speed up early expansion and alter the sound horizon. New neutrino physics, dark radiation, primordial magnetic fields, varying fundamental constants or modified recombination histories can produce related effects.

Unfortunately for simple solutions, the CMB and Big Bang nucleosynthesis already constrain much of this parameter space. ACT DR6 found no significant evidence for self-interacting dark radiation or early variation in the fine-structure constant or electron mass. Its extended-model analysis concluded that the tested models were not favored over ΛCDM.

So early-Universe new physics is not excluded. The point is that precision data have turned a once-large playground into a narrow target.

Late-Time Dark Energy and Modified Gravity

Another idea is to change the recent expansion history. If dark energy evolves instead of behaving as a cosmological constant, or if gravity differs subtly from general relativity on cosmic scales, perhaps local and early-Universe measurements can be reconciled.

The problem is that supernovae and BAO already map the relative late-time expansion history very well. Changing H(z) enough to lift H₀ tends to distort those distances. This is one reason reviews increasingly emphasize that purely late-time fixes are difficult.

DESI’s hints of evolving dark energy are therefore fascinating but should not be confused with a demonstrated Hubble-tension solution. A Universe with dynamic dark energy could be real and the Hubble tension could still require a different explanation.

The Supernova Absolute-Magnitude Problem

There is another way to state the puzzle that is often more revealing than quoting H₀. Type Ia supernovae map relative distances extremely well, but they do not know their own absolute luminosity. Local distance ladders calibrate that absolute magnitude using nearby stars. Inverse ladders calibrate cosmic distances using the early-Universe sound horizon and BAO.

The two calibration routes imply different absolute scales. Research combining DESI, supernovae and sound-horizon information emphasizes a degeneracy between the supernova absolute magnitude and the primordial ruler. In other words, solving the tension is not just about changing a number called H₀: the entire absolute calibration system has to become mutually consistent.

Why the Standard Model Is So Difficult to Replace

Calling this a “crisis in cosmology” can create the wrong picture. ΛCDM is not hanging on because cosmologists are reluctant to abandon it. It became standard because six parameters reproduce an extraordinary range of observations — the CMB, BAO, large-scale clustering and much of the expansion history — with remarkable economy.

The model is also conceptually incomplete. We do not know the particle identity of cold dark matter. We do not understand why vacuum-like dark energy has its observed tiny density. Inflation and the origin of primordial fluctuations sit outside the minimal six-parameter late-time model. So physicists already know ΛCDM is not a final theory of everything.

The Hubble tension therefore asks a sharper question than “Is ΛCDM imperfect?” We already know it is incomplete. The question is whether one of those missing pieces has finally become visible in precision cosmological data.

What Would Actually Count as a Breakthrough?

Not another decimal place from the same calibration chain. A breakthrough would be convergence among methods whose systematic errors are genuinely unrelated.

If standard sirens, strong lenses, megamasers and multiple independent JWST stellar indicators converge near 73 while ACT, future CMB experiments and BAO remain near 68 under ΛCDM, the case for missing cosmological physics would become much harder to avoid. If the independent local methods drift toward 68–70 as samples improve, the story would instead point toward calibration or population effects in the highest local determinations.

There is also a less cinematic possibility: no single side wins. Several small calibration effects and modest model changes could meet in the middle. The famous 67-versus-73 split may eventually turn out to be the visible sum of more than one problem.

The Experiments That Could Decide the Case

JWST will continue expanding samples in which Cepheids, TRGB and JAGB stars can be measured in the same galaxies. That matters because comparing methods within identical hosts removes an entire layer of sample mismatch.

Gaia improves the parallax foundation of stellar calibration. DESI continues to sharpen the BAO and full-shape expansion history. ACT’s final data are now a major independent CMB reference, while future ground-based CMB surveys will push polarization and lensing measurements further. Strong-lensing programs are adding systems and better stellar kinematics.

Gravitational-wave astronomy may ultimately provide the cleanest psychological test because its distance scale comes from physics utterly different from stellar candles. The present standard-siren errors are large, but hundreds or thousands of well-characterized events could make them one of the most important arbiters of H₀.

The answer may come not from one better measurement, but from several independent ones. JWST distance indicators, gravitational-wave standard sirens, CMB observations and enormous galaxy surveys are approaching the problem from different directions.

So, Is Our Model of the Universe Wrong?

It could be. But cosmology has not earned that conclusion yet.

As of 2026, the strongest form of the mystery is still alive. High-precision Cepheid-based local measurements remain near 73, and JWST has not uncovered the large crowding bias that could have neatly explained them. ACT and DESI, interpreted within ΛCDM, remain near 68. Other JWST calibrators and independent methods sit between the camps or still carry uncertainties wide enough to include both.

That leaves three broad possibilities: an unidentified systematic survives somewhere in the measurement chains; ΛCDM is missing physics that changes the absolute calibration between the early and late Universe; or several smaller effects are masquerading as one clean discrepancy.

And that brings us back to what makes the Hubble tension so unsettling. We can build one cosmic ruler outward from nearby stars. We can build another forward from light released when the Universe was only about 380,000 years old. Each chain is supported by physics that succeeds spectacularly elsewhere. Yet when both rulers reach the present-day Universe, they do not mark the same scale.

If the gap disappears, it will be a lesson in how one-percent systematics can survive years of precision astronomy. If it remains after genuinely independent tests mature, those few kilometers per second per megaparsec may become evidence that the standard cosmological model is missing something fundamental.

Either way, the disagreement is doing exactly what a good scientific mystery should: forcing our best measurements — and our best theory — to face one another.

Key Numbers to Remember

Planck 2018 + base ΛCDM: about 67.4 ± 0.5 km/s/Mpc.

ACT DR6 + CMB lensing + DESI DR2 in ΛCDM: 68.43 ± 0.27 km/s/Mpc.

SH0ES 2022 Cepheid–SN ladder: 73.04 ± 1.04 km/s/Mpc.

SH0ES expanded JWST/HST Cepheid analysis (2025): 73.49 ± 0.93 km/s/Mpc.

SH0ES Cepheid + TRGB combination (2025): 73.18 ± 0.88 km/s/Mpc.

CCHP JWST three-method combination (2024): 69.96 ± 1.05(stat) ± 1.12(sys) km/s/Mpc.

Megamaser Cosmology Project: 73.9 ± 3.0 km/s/Mpc.

TDCOSMO 2025 time-delay lenses + Pantheon+: 72.1 +4.0/−3.7 km/s/Mpc.

FAQ: Hubble Tension Explained

What is the Hubble tension?

The Hubble tension is the persistent disagreement between high local determinations of the present expansion rate, often around 73 km/s/Mpc, and lower values around 67–68 inferred from early-Universe and BAO data within the standard ΛCDM model.

Why do Planck and SH0ES get different Hubble constants?

They use fundamentally different calibration routes. SH0ES builds a local distance ladder from geometric anchors to Cepheids and Type Ia supernovae. Planck measures the early-Universe CMB and infers today’s H₀ through ΛCDM.

Did JWST solve the Hubble tension?

No. JWST strongly reduced concerns about Cepheid crowding in SH0ES measurements, which continue to give a high H₀. But another JWST program using Cepheids, TRGB and JAGB obtained a lower combined value near 70 with larger uncertainties. Webb has clarified the problem rather than ended it.

Is the Hubble tension proof that ΛCDM is wrong?

No. It is a serious stress test, not proof. Hidden systematic errors remain possible, and proposed new-physics models must fit many other observations that ΛCDM already explains successfully.

What is early dark energy?

Early dark energy is a proposed temporary energy component before recombination. By increasing the early expansion rate, it can shrink the sound horizon and allow a larger inferred H₀. Current data constrain simple versions strongly, and no EDE model is established as the solution.

Could a giant cosmic void around us explain the tension?

A local underdensity can slightly affect local expansion measurements, but current evidence indicates that a void large enough to explain the full discrepancy is not a viable dominant solution.

What could finally resolve the Hubble tension?

The strongest resolution would come from independent techniques with unrelated systematics, especially larger gravitational-wave standard-siren samples, more time-delay lenses and megamasers, and direct JWST comparisons of multiple stellar distance indicators in the same supernova hosts.