Search
Results for “zk-STARK vs zk-SNARK”
zk-STARK
A zk-STARK is a kind of zero-knowledge proof that needs no secret setup ceremony and relies on hash functions. It lets anyone check that a large computation was done correctly far faster than redoing it, at the cost of proofs larger than zk-SNARK proofs.
What it means for you. Because there is no secret setup, there is no hidden value someone could have kept to forge proofs; you still rely on the proving code and the program being proven being correct. Larger proofs cost more to verify on a base chain, which is part of what systems using them pay.
Sources: Ben-Sasson et al., Scalable, transparent, and post-quantum secure computational integrity (IACR ePrint 2018/046), ethereum.org: Zero-knowledge proofs · checked 4 October 2026
zk-SNARK
A zk-SNARK is a kind of zero-knowledge proof that is very small and quick to check: it convinces anyone that a computation was done correctly without revealing the private inputs or redoing the work. Many designs need a one-time setup ceremony whose secret values must be destroyed afterwards.
What it means for you. Systems built on zk-SNARKs, such as some rollups and privacy tools, ask you to rely on their setup ceremony having had at least one honest participant and on the proving code being correct; a flaw in either could allow false proofs. Whether a system needs a setup, and who took part, is a concrete question you can check.
Sources: ethereum.org: Zero-knowledge proofs, Groth, On the Size of Pairing-based Non-interactive Arguments (IACR ePrint 2016/260) · checked 4 October 2026