Recently, the market has been moving sideways, with not many hotspots, and the sentiment of retail investors cannot be described as exuberant. However, it is precisely during such “directionless” times that smart money often starts to move quietly. Especially in those old narratives that have been discussed repeatedly and were even once thought to be “untradeable.”

On April 29 this year, the privacy blockchain 0xMiden announced it has secured $25 million in seed round financing, led by a16zcrypto, hackVC, and 1kxnetwork, with participation from Finality Capital Partners, Symbolic Capital, and individual investors including Avery Ching (CEO of Aptos Labs) and Rune Christensen (founder of MakerDAO).

0xMiden focuses on the concept of “edge blockchain”, with the goal of allowing applications to choose an open and transparent path when processing transactions, and can also switch to privacy mode with one click, so as to achieve privacy protection without sacrificing scalability.

!

Seeing it being surrounded by big players in the ZK and privacy narrative, I am a bit curious: what key point did this project hit that attracted so much capital? How is 0xMiden different from projects like Aztec and Zcash that have long been focused on privacy?

To clarify this issue, we need to understand one thing: the current blockchain is actually very “non-private”, but if you blindly pursue privacy, you will sacrifice scalability and compliance.

So in the direction of privacy chains, although there have been voices in the past few years, there has never been a solution that is “usable, scalable, and not subject to one-size-fits-all regulation.”

Now, 0xMiden attempts to restart this dilemma from a different perspective.

A Privacy Chain That Can Run Without Internet Connection

0xMiden claims to be an “edge blockchain”, which sounds a bit like edge computing. The core idea is that smart contracts are no longer entirely handled by the chain; instead, users execute them locally on their own devices, and once computed, package them into a ZK proof and broadcast it to the chain for verification.

In other words, each device is like a “small blockchain node”, calculating and proving by itself, only telling the chain the results and letting the chain verify it for you. In this mode, the role of the chain is no longer a “coolie”, but more like a “referee”, you say that you are right, then you have to come up with a verifiable proof.

This model has several core changes, all executions are completed locally, with very little burden on the chain, making scalability naturally strong. The generated ZK proof is lightweight, unlike the super heavy batch proofs of zkRollup, suitable for terminal devices to handle on their own. By default, it is private, but developers can choose which computations to make public and which to keep confidential, allowing for flexible switching to adapt to different scenarios.

In the end, what it wants to create is not a large and comprehensive Layer 1, but a new way of interaction between chains and devices, running smart contracts like a portable wallet.

This has some obvious differences from the popular privacy chains on the market today. For example, Aztec focuses on “ZK-rollup + encrypted EVM”, which still rolls up all user transactions for processing, and the sequencer is also very centralized, essentially following the Layer 2 approach. Fhenix uses TEE + homomorphic encryption, leaning more towards hardware trust + encrypted storage, with ZK not being the main character.

0xMiden, from a fundamental logical perspective, has taken a “ZK-first + local computation” approach, emphasizing flexibility, lightweight, and terminal friendliness.

In simple terms, the execution of 0xMiden is “decentralized”, and it can even be executed offline, generating proofs offline, and then uploading the results uniformly. Your mobile phone, browser, and AI chips may all become small nodes running smart contracts in the future, and ZK helps ensure the correctness of calculations, while the chain only verifies the results.

Why are the big shots willing to spend money?

As early as 2021, a16z invested in the early project of 0xMiden founder Bobbin Threadbare, which was then part of Polygon Miden. The participation of a16z in the $25 million seed round financing on April 29, 2025, highlights its long-term confidence in the Bobbin team.

On the team side, 0xMiden was founded by Bobbin Threadbare, Dominik Schmid, and Azeem Khan, all of whom previously served on Meta’s blockchain team and possess extensive experience in zero-knowledge proofs and blockchain development.

0xMiden was originally a project developed under Polygon Labs and spun off into a separate entity on March 31, 2025, and the financing was considered a “team restructuring + independent financing”. After the spin-off, 0xMiden focused on the Ethereum zk-Rollup privacy solution, and CryptoRank served as the incubator, providing funding and resource support, but the technical development was mainly completed by the team independently.

0xMiden is the third project to be spun off from Polygon, following Avail and Privado ID. Sandeep Nailwal, co-founder of Polygon Labs, stated in a release: “It plans to airdrop about 10% of its native token to POL holders and stakers, directly rewarding those who help develop its ecosystem.”

From an investment perspective, the appeal of 0xMiden lies in its ZK native computing model, which represents the future trend of ZK. Unlike traditional Rollups, it is not a replacement for chains but is developed as a privacy plugin, providing flexible privacy options.

Project Status

At the same time, Miden has launched useful development tools that allow developers to start writing real-world smart contracts, away from projects that are only in the concept stage. What’s more, it is designed to support “privacy by default, optional public”, and there is also some room for compliance.

The mainnet has not been launched yet, but the VM and SDK have been open-sourced on Github. Their provided roadmap is as follows:

·Miden VM open source (completed)

·Developer Toolkit SDK Release (Completed)

·Testnet launch (expected Q3 2024)

·Mainnet launch (as early as early 2025)

Some developers are already using Miden VM to write demo contracts, such as local voting, device log verification, and in-game asset operations. However, to truly “break the circle,” we may have to wait for a “killer application scenario” like a “local version of Friend.tech” or an “AI model running market.”

Why has privacy blockchain become popular again?

This narrative is not new, but it has indeed seen a bit of a resurgence recently, mainly for three reasons.

First, regulatory pressure is continuously rising. Globally, regulations on crypto identity and asset flow are becoming increasingly stringent, whether it is on-chain KYC, wallet blacklists, or exchange compliance, all of which are constantly squeezing the living space of “absolute anonymity.” Instead, privacy demands that are institution-facing, compliance-friendly, and within controllable scopes are heating up, especially in enterprise-level scenarios, where selective privacy has become a necessity.

Secondly, ZK technology has finally become “usable”. For the past few years, ZK has remained in the stage of “technical show-off”, being expensive, slow, and difficult to develop. However, with the maturity of technologies like STARK and SNARK, the costs of proof generation and verification have continuously decreased, transforming ZK from a “theoretical future” into a “practical reality that can be run effectively.”

Thirdly, with the rise of AI blockchain applications, more and more scenarios such as AI computing, data on-chain, and edge inference are being implemented. The demand that was originally limited to transaction privacy is beginning to expand to “data privacy” and “model privacy.”

The privacy requirements for this type of application are no longer “full-chain black box”, but rather “local computation + selective encryption”, which is known as local privacy.

On one hand, established privacy projects like Monero and Zcash only support simple transactions, and on the other hand, they are under close regulatory scrutiny. The path of promoting “absolute anonymity” is becoming increasingly difficult.

So, new opportunities have arrived, not to create an absolute black box, but to implement “local computation + selective privacy.”

Will this round of privacy narrative start from “local”?

The direction of privacy chains is not likely to explode in the short term, but it has always lacked a truly viable solution.

Aztec and Fhenix are still working on “on-chain privacy,” while 0xMiden wants to bring this narrative to “off-chain” devices, operating independently and selectively on-chain.

This angle is actually the next natural evolution direction of ZK, and it aligns with future trends such as “AI + blockchain” and “data sovereignty”. If this story can be articulated well and the product can truly be implemented, then 0xMiden may become the next “privacy computing infrastructure” worth paying attention to.

Of course, the premise is that it can really hold out until the day the mainnet goes live.

：