What is a zk-STARK?
- Author
- CRYPTO PORT Editorial
- Published
- Updated
- Reading time
- 6 min
In short
A zk-STARK needs no trusted setup at all. Its security rests only on hash functions, which also makes it a plausible candidate for resisting quantum attacks. The trade-off is proof size: STARK proofs are substantially larger than SNARK proofs, so posting them on-chain costs more data.
Key points
- No trusted setup — the setup is fully transparent
- Security relies only on hash functions
- Proofs are larger than zk-SNARK proofs
- Used by Starknet and StarkEx
Definition
Short for Zero-Knowledge Scalable Transparent ARgument of Knowledge: a proof system whose setup requires no secret (transparent) and which scales to very large computations.
The defining difference is the 'transparent' part. Public parameters are derived from public randomness alone, so there is no ceremony to run and no participant whose deletion of a secret you must trust. That removes an entire trust assumption.
The security base differs too. zk-SNARKs lean on hard problems over elliptic curves; zk-STARKs lean only on hash collision resistance. Shor's algorithm threatens elliptic curves on a quantum computer, whereas quantum attacks on hashes only speed up search, which is why STARKs are seen as the better long-term bet.
The cost is size. Where a SNARK proof is a few hundred bytes, a STARK proof can run to tens of kilobytes, and on Ethereum data is cost. StarkEx and Starknet answer this by amortising one proof over a very large batch of transactions so the per-transaction share stays small.
Watch out for
- · Quantum resistance here refers to the proof system, not to the chain's signature scheme
- · Larger proofs mean higher data costs, which is why some projects still choose SNARKs
- · This is young tooling; check audit history and production track record