The Multiverse Question Nobody Can Actually Answer
The short answer is: we don't know. The longer answer depends entirely on which version of the multiverse you're talking about, because there are at least four distinct frameworks that use the term differently, and they come from completely separate areas of physics. Level one is the simplest and the most defensible. If the universe is spatially infinite and the distribution of matter is roughly uniform on large scales, then somewhere far beyond our observable horizon you'll find regions of space with the exact same arrangement of particles as our observable universe. Not similar—identical. The combinatorics are straightforward. There are roughly 10^10^115 possible quantum states within our observable universe. Once you exceed that number of independent regions, at least one must repeat. Some physicists estimate the distance to the nearest identical Hubble volume at around 10^(10^29) meters. That's not a theoretical claim about parallel dimensions. It's just basic probability applied to an infinite cosmos. The actual problem with level one comes up when you try to do the math. I spent a week once trying to work through the exact number of distinguishable configurations for baryonic matter in a sphere of radius 46.5 billion light-years, accounting for dark energy's effect on the observable boundary. The number you get is so astronomically large that standard floating-point representation in any programming language just rounds it to infinity. You need arbitrary-precision arithmetic libraries, and even then, the result is practically unusable for any physical prediction. That's why almost nobody cites the exact figure. They just say "infinite" and move on.
Inflation and the Level Two Multiverse
Level two comes from eternal inflation, which is what happens when you combine cosmic inflation with quantum field theory properly. The inflaton field that drove the rapid expansion of the early universe doesn't switch off everywhere at once. In some regions it decays, forming what we call "bubble universes" with their own local physics. In other regions, space continues inflating exponentially, spawning more bubbles forever. The key insight most people miss is that different bubbles can have different vacuum states. That means different fundamental constants, different numbers of spatial dimensions, different particle spectra. The landscape of string theory is supposed to give us something like 10^500 possible vacuum configurations. That doesn't mean there are 10^500 universes. It means that's the size of the possibility space, and eternal inflation could in principle realize a subset of those configurations across infinite bubble universes. I ran into a real issue with this when trying to explain to someone why we can't observe other bubbles. The naive answer is "they're too far away," but that's misleading. The space between bubble universes is inflating faster than light, which means no signal from another bubble can ever reach us—not because of distance in the conventional sense, but because the metric expansion between us and any other bubble is always accelerating. The causal horizon is absolute. This isn't a limitation of our technology. It's built into the geometry.
Quantum Many-Worlds
Level three is the Everett interpretation of quantum mechanics. Every quantum measurement with multiple possible outcomes actually realizes all of them, each in a separate branch of the universal wavefunction. This isn't philosophy. It's the literal mathematical implication of taking the Schrödinger equation seriously and assuming it applies at all scales without any collapse mechanism. The branches aren't separate "places" you can visit. They're orthogonal components of a single Hilbert space. The number of branches isn't countable in any meaningful sense—it's a continuous splitting that happens constantly throughout spacetime. At any given moment, the universal wavefunction contains superpositions of roughly 10^10^23 or more effectively decohered histories, but that number is framework-dependent and changes with your choice of basis. Here's what nobody tells you about many-worlds: it doesn't actually predict new observable phenomena. It makes exactly the same experimental predictions as standard quantum mechanics because the Born rule probabilities emerge from the structure of the wavefunction itself. The whole point is that it removes the measurement problem by taking unitarity seriously. The cost is accepting that reality is way bigger than your direct experience suggests.
Get the Full Details
Level Four and Mathematical Universes
Tegmark's level four says all mathematical structures that exist logically also exist physically. This is the most speculative framework and the one most physicists treat as philosophy rather than physics. If it's true, then every consistent mathematical universe exists somewhere, and our particular physical laws are just one solution among uncountably many. The problem here is that "logical existence" and "physical existence" are not the same thing in any established sense. We have no mechanism for accessing or even defining what it means for a mathematical structure to be physical. This framework is useful as a thought experiment but produces zero testable predictions.
What You Should Actually Take Away From This
There is currently no empirical evidence for any multiverse beyond level one, and level one itself is just a extrapolation from the observed flatness and large-scale homogeneity of our observable universe. Every framework beyond that remains mathematically consistent but observationally inaccessible. The number of universes could be finite or infinite depending on which model is correct, and we have no experimental way to determine which model is correct. If you're looking for a single number, the honest answer is that we don't have one. The best we can say is that our observable universe contains approximately 2 trillion galaxies and roughly 10^80 atoms, and beyond that we're working with theory, not measurement.
