Most lists of “ZK grants” are lists of every grant program in crypto. That does not help a team building a prover, a zkVM or a circuit library, because most ecosystem programs fund applications on their own chain, not the cryptography underneath. The useful question is narrower: which programs say, in their own stated focus, that they fund ZK, cryptography or privacy work, and what does each one want in return?
This article answers it from our Web3 Grants Tracker: 27 programs, last verified on 27 September 2026. Tickets are quoted as the programs state them; where there is no fixed amount we write “varies” rather than guess. Terms change, so confirm on each program’s own page before you apply.
Four programs that name ZK, cryptography or privacy
- The Ethereum Foundation’s Ecosystem Support Program. Focus: public goods, infra, research, cryptography/ZK. Ticket varies, no equity, any stage. Applications are rolling, and next to grants the program lists wishlist and RFP items. The focus line leads with public goods, so pitch work the wider Ethereum ecosystem can use, not a product only your company benefits from. Read the wishlist and RFPs first; if one matches your work, write the application against it.
- Aleo Network Foundation developer grants. Up to $100K, no equity, any stage. Focus: privacy-first and zero-knowledge apps and tooling on Aleo. Grants are milestone-based, open source is required, and the program lists initial feedback within one week. It wants something built on the Aleo network, so this is not a general-purpose prover grant. The application asks for a project overview, the team’s background and a proposal with a development roadmap.
- Zcash Community Grants. Focus: the Zcash ecosystem, meaning privacy tech, wallets and infrastructure. Ticket varies; the tracker lists major grants and roughly 100K ZEC available through November 2028. No equity, any stage. It is funded from the Zcash Dev Fund and decided by an elected five-person committee, which also accepts RFP ideas. The stated focus is the Zcash ecosystem, so lead with what your work does for Zcash, not with proof-system novelty.
- ZK Prime Capital Grants. $5K–$75K, no equity, idea or prototype stage, a reply within 5 business days. Focus: on-chain AI agents, DeFAI and ZK infrastructure. Inside ZK that means proving systems and zkVMs, ZK for AI, developer tooling and privacy rails. This is a VC grant program, so the question is whether the work can become a company: if the milestone ships, there is an optional $50K–$250K round, incubation and a token launch.
Programs that fit by stack, not by name
The Starknet Foundation does not name ZK in its focus line (Starknet apps, tooling, infra), but Starknet is built on STARK proofs, so tooling and infrastructure work there is often ZK work in practice. Seed Grants go up to $25K in STRK for an MVP or proof of concept; Growth Grants run from $25K to $1M in STRK for production products.
Three public-goods programs can fit an open-source ZK library once it has shipped and people use it. Octant allocates community-directed ETH in recurring epochs to live Ethereum public goods, including infra, security, research and tooling. Optimism Retro Funding pays retroactively for impact already delivered on the Superchain and has a Dev Tooling mission; you enroll through OP Atlas. Gitcoin runs seasonal quadratic-funding rounds for open source and public goods. Octant and Retro Funding are listed for live work only; Gitcoin lists any stage, but quadratic matching follows how many people back you.
Nearly every other program on the tracker funds its own ecosystem. A ZK component is fundable there only when it makes that chain more useful, so read the focus line first. For programs outside the tracker, take the ticket and terms from their own pages.
Idea stage vs prototype: how to sequence applications
- Idea stage: a design, no working code. Start where an idea is explicitly in scope and the reply is fast. ZK Prime lists idea or prototype stage and replies within 5 business days. The Ecosystem Support Program, Aleo and Zcash Community Grants list any stage, so an early application is allowed, but it needs one concrete milestone. For ZK work that is usually a public benchmark, or a minimal circuit and verifier on testnet. Leave Starknet Seed until you have the MVP or proof of concept it is designed for, and leave retroactive and quadratic rounds for later, when the work is delivered and has backers.
- Prototype: a prover, circuit library or verifier that runs, with numbers. Now the measured result carries the application. Apply to the stack you built on: Starknet Seed if it runs on Starknet, Aleo if it is built on Aleo, Zcash Community Grants if it serves Zcash. Send the chain-neutral part, such as the library or the benchmark suite, to ESP. Once the code is open, shipped and in use, add Octant, Retro Funding and Gitcoin.
Two rules hold at both stages. First, apply to rolling programs with short reply times before round-based ones, so that an early yes becomes evidence in the later applications. Second, never ask two programs to pay for the same milestone: give each its own deliverable and say what else is pending. Our guide to applying to several grant programs at once covers how to split the work and how to disclose it.
What reviewers check in a ZK application
- Benchmarks someone else can reproduce. State proving time, memory, proof size and verification cost, on named hardware, for a named circuit or program size, against a named baseline at the same security level. Publish the script that produces the numbers. At idea stage, state the target and how you will measure it.
- Soundness assumptions, in plain words. Name the proof system, the cryptographic assumptions it rests on, whether there is a trusted setup and who runs it, and the security level you target. Separate what is proven from what is conjectured. Say what has been audited or reviewed, what has not, and how you test for under-constrained circuits. Reviewers accept “not yet audited”; they do not accept discovering it themselves.
- The open-source licence. Aleo lists open source as required, and public-goods rounds exist to fund open work. Put a licence file in the repository before you apply, name the licence in the application and flag any dependency that ships under different terms. If part of the stack will stay closed, say which part and why, then check each program’s terms to see whether that is acceptable.
- A path to deployment. Say who integrates the component and what they need from it. In our own program a paper alone is not fundable; “$40K gets the prover benchmark public and a Noir library on testnet in 10 weeks” is.
What we fund in ZK, and what we usually pass on, is laid out on our ZK infrastructure grants page. Our one-page application template on GitHub works for any of the programs above.
Building a prover, a zkVM or ZK tooling? Pick the two or three programs from the tracker that match your stack and stage, and send us the same one-pager. Apply with your project ▸