Why people compare them
Both are sold for tendon, ligament and muscle recovery, often together as a "healing stack". Both have plausible mechanisms in animal models — BPC-157 via angiogenic and nitric-oxide pathways, thymosin beta-4 via actin sequestration and cell migration. And both are conspicuously absent from clinical practice, which is the first clue about the state of the evidence.
The important asymmetry
BPC-157's literature is large but narrow: hundreds of animal papers, a majority from a single research group in Zagreb, and until recently no human trials at all. TB-500's literature is borrowed: the papers cited for it are almost always about full-length thymosin beta-4, a different molecule that has been through early human trials for dermal wounds and eye disease. Whether the fragment sold as TB-500 behaves like the parent peptide in humans has not been tested.
What would settle it
A randomised, placebo-controlled human trial in a defined injury with a functional endpoint. As of our last update we can find no such trial with posted results for either compound. Until then, "which is better?" has no evidence-based answer — and "does either work in people?" is still open.