To assess a project’s progress, I don’t care what the main net narrative says—I look at its GitHub repository. In the code of @Dusk , there’s a very honest signal: in issues on the documentation repository, it states that after adopting a modular architecture, it is recommended that developers use DuskEVM, but most of the existing developer documentation still talks about DuskVM. In other words, the product direction has already shifted, but the documentation and code haven’t fully caught up yet.

The official GitHub repositories published by Dusk also confirm this mismatch. dusk-network/rusk is a reference implementation; it includes core modules such as dusk-vm, dusk-core, and PLONK zero-knowledge proofs. However, the standalone rusk-vm repository—which is closer to native capabilities—doesn’t seem to be generating much buzz; the star count and update signals look much quieter than the main story. Meanwhile, the official documentation has already placed DuskEVM’s quickstart, Solidity, and the Ethereum toolchain in the most convenient locations. A project that markets native privacy and WASM smart contracts, yet in practice steers new developers toward the EVM path—this is, in itself, a silent statement. #dusk

I haven’t treated the star counts as a conclusion yet, because code activity depends on more than stars: commit frequency, number of contributors, and issue-closure rate. But when these fragments are put together, the direction becomes very clear: $DUSK ’s focus is on migration, and the native VM portion feels more like a capability meant to be retained long-term rather than the primary developer entry point being pushed right now. The “mainnet coming” in the official announcement needs a corresponding cadence of code delivery to back it up—not just technical documentation moving directories.

The conclusion is: at the moment, the publicly visible code signals are not enough to support the claim that “two execution environments are equally mature.” DuskEVM is clearly the more active, more heavily promoted direction, while DuskVM looks more like a reserve that hasn’t been fully refined yet. Real progress should be measured by whether the issues across the next few milestones are being steadily closed—not by the fact that a few more concept pieces have appeared on the homepage.