arguably we wouldn't have JWST without Hubble, so cost adjusted they're about the same, then again, there is probably some "prior art expensive project" associated with Manhattan, but I didn't check :)
As for the Manhattan Project, it was sufficiently close to the first direct theoretical and applied theory and proofs of sustainable nuclear chain reactions (roughly a decade or less following each), and a lack of understanding of the corresponding risks (which greatly increase costs) that there simply wasn't time to have spent all that much money.
By contrast, Hubble and JWST are both late-stage, highly-evolved technologies, pushing the engineering envelope in many dimensions simultaneously, all of which tends to increase costs.
See for example the ELT (extremely large telescope), an Earth-based instrument currently under construction in Chile as part of ESO (European Southern Observatory). Tom Scott's 'splainer video on the project explains how costs risk exponentially with increased size for numerous reasons.
1945 cost of the Manhattan Project, $1.89 billion,[1] or $32.9 billion 2024 per <https://www.usinflationcalculator.com/>.
2016 cost of the JWST: ~$10 billion,[2] or $13.1 billion 2024 adjusted for inflation.
That's a smaller multiple for the Manhattan Project than I'd have expected, but it's still comfortably more expensive than the JWST.
________________________________
Notes:
1. <https://en.wikipedia.org/wiki/Manhattan_Project#Cost>
2. <https://en.wikipedia.org/wiki/James_Webb_Space_Telescope>