ArXiv preprint claims axion early dark energy model raises Hubble constant to 71.6, easing Hubble tension

A new arXiv preprint argues that a specific early dark energy model, fitted to data from the cosmic microwave background, DESI and JWST, can push the inferred Hubble constant up to 71.58 ± 1.05 kilometers per second per megaparsec and largely erase the long-running “Hubble tension.” But the result is unreviewed, depends on a particular model choice, and does not amount to a settled resolution of the dispute.

The paper, “Resolving the Hubble Tension in the Early Dark Energy Framework with JWST and DESI Data,” was posted June 17, 2026, as arXiv:2606.19090 v1 in astro-ph.CO. Its authors are Guo-Hong Du, Tian-Nuo Li, Lu Yin, Sheng-Han Zhou, Hao Wang, Jing-Fei Zhang and Xin Zhang. In the paper’s abstract, the authors say they analyze cosmic microwave background data from Planck, ACT and SPT, along with baryon acoustic oscillation measurements from DESI and ultraviolet luminosity-function observations from JWST. Within what they call a canonical axion early dark energy framework, they report H0 = 71.58 ± 1.05 km s−1 Mpc−1, reducing the tension to the 1.0-sigma level. They also say their model fits the selected data better than standard Lambda-CDM, reporting Δχ²_tot = −18.26 and ΔDIC = −11.89 in their analysis.

The Hubble tension refers to the persistent mismatch between estimates of how fast the universe is expanding today. Using the standard Lambda-CDM cosmology, Planck 2018 inferred a Hubble constant of about 67.4 ± 0.5 km s−1 Mpc−1 from the early universe. A widely cited local-distance-ladder measurement from the SH0ES team led by Adam Riess found about 73.04 ± 1.04 km s−1 Mpc−1. If the gap proves real and not the product of hidden errors, it would point to new physics beyond the standard cosmological model.

Early dark energy is one proposed way to do that. In broad terms, it adds a short-lived burst of extra energy in the early universe, around the era when the cosmic microwave background was formed. That can shrink the “sound horizon,” the standard ruler used in cosmic microwave background and galaxy-survey fits, which in turn allows those fits to accommodate a higher present-day Hubble constant.

JWST and DESI matter here because both have become important inputs in current tests of cosmology. In some analyses, JWST has found unexpectedly abundant bright galaxies at very high redshift, while DESI has produced major baryon acoustic oscillation measurements that are central to checking both Lambda-CDM and alternatives to it. The new preprint argues that combining those data with cosmic microwave background observations strengthens the case for this version of early dark energy.

Still, the claim comes with major caveats. The paper is a preprint on arXiv, and the arXiv record does not indicate publication in a peer-reviewed journal. Early dark energy has been studied for years, and past results have been mixed depending on the exact model assumptions and on which cosmic microwave background datasets are included. More broadly, models that raise H0 have often drawn scrutiny because they can shift other cosmological parameters, including S8, which tracks matter clustering, so lifting the Hubble constant alone does not end the debate.

For now, the narrow takeaway is that the preprint adds a fresh, specific claim to one of cosmology’s most closely watched disputes: under one axion-like early dark energy model and one combination of CMB, DESI and JWST data, the Hubble tension can be reduced sharply. Whether that holds up under peer review and independent reanalysis is the real test.