Post-quantum · MEV-free · Sub-second · Commodity-decentralized
A Layer 1 blockchain built greenfield to ship four properties as defaults at genesis — none of which any production chain combines today:
- Post-quantum cryptography — FALCON-512 signatures, Kyber-768 threshold encryption, Poseidon2 + Blake3 hybrid hashing. No pre-quantum primitive on any consensus or account path.
- MEV resistance by structure, not policy — threshold-encrypted mempool + commit-before-reveal ordering + DAG consensus. Sandwich attacks, front-running, and proposer extraction are not auctioned or mitigated — they are structurally impossible.
- Sub-second finality — Mysticeti-style DAG consensus with ~500ms median commit, 85-of-128 FALCON quorum certificates.
- Commodity-hardware decentralization — full nodes and validators awaiting committee selection run on 8c / 16GB. Equal voting power within the active committee; single-tier 10,000 PYDE stake minimum, uniform-random committee selection per epoch.
The execution layer is WebAssembly via wasmtime, with Cranelift
ahead-of-time compilation and a hybrid parallel scheduler combining
static access lists with Block-STM speculation. Smart contracts can
be authored in Rust, AssemblyScript, Go (TinyGo), or C/C++ —
whatever language fits the team — and bundled by the otigen
developer toolchain, which handles language detection, build
invocation, state binding generation, and deploy submission.
Pyde is pre-mainnet. Architecture design is complete. The protocol has gone through two clean pivots, both honest, both for the right reasons:
- Consensus pivot — from an in-house HotStuff variant (whose persistent wedges and stalls at 400ms slot timing motivated a clean rebuild) to Mysticeti-style DAG consensus.
- Execution pivot — from a custom virtual machine and a domain-specific language (Otigen) to WebAssembly via wasmtime. The Otigen name lives on as the developer toolchain. The full story is in pyde-book/src/preface/pivot.md.
| Component | State |
|---|---|
| Architecture design | Complete |
| WASM execution layer (wasmtime + Cranelift) | Foundation in place, integration in flight |
| State layer (JMT, dual-hash Blake3+Poseidon2) | In place, dual-hash wiring in flight |
| Mysticeti DAG consensus | Rebuild in progress post-pivot |
| Threshold cryptography (PQ) | Research-grade — bleeding-edge |
| Network protocol | Existing; libp2p + QUIC migration in flight |
| Otigen developer toolchain (WASM-era) | Specification in flight, scaffold to follow |
| Performance harness | Not yet built (mandatory before any TPS claim) |
Mainnet ships when the implementation is complete, audited, and validated by an incentivized testnet. No public schedule. No external TPS claim without harness evidence — the "claim 1/3 of measured peak" discipline.
The comprehensive technical reference is the Pyde Book — read it
at book.pyde.network (source lives at
pyde-net/pyde-book):
- 📘 Pyde Book — full architecture: 20 chapters covering the VM, state model, consensus, cryptography, MEV protection, gas/fee model, account model, networking, governance, security, and the phased launch plan. Includes companion specs for the threat model, slashing, validator lifecycle, state sync, chain halt + recovery, and the performance harness.
Most repositories in the pyde-net org
are private during pre-mainnet engineering — access on request.
Repositories will be opened publicly as the implementation stabilises:
the post-pivot engine workspace (yet to be re-cut), the otigen
developer toolchain (WASM-era, yet to be bootstrapped), the Rust and
TypeScript SDKs, the block explorer, and the PIP registry. The
pre-pivot otic compiler, wright toolchain, and the original
Otigen language book are preserved as historical artifacts in the
archive repo and in their
own retired-status repos.
Pyde is built by zarah, making the technical choices and shipping the work. The consensus pivot from HotStuff to Mysticeti DAG and the execution pivot from a custom VM to WebAssembly both reflect an explicit preference for designing from a clean foundation over patching accumulated technical debt — and for killing darlings honestly when the evidence says they should be killed.
Engineering rigor is the project's discipline. Every claim in the book is grounded in either code that's running or an explicit "this is the designed behavior; the code is in flight." Performance numbers come from the harness — never from microbenchmarks or local devnet measurements — under the "claim 1/3 of measured peak" rule.
This is not vaporware. It is also not yet a shipped chain. The distinction matters.
- 📜 Contributing — how to propose changes, PIP process, engineering standards
- 🔒 Security policy — vulnerability disclosure, scope, safe harbor
- 🤝 Code of Conduct — community standards (Contributor Covenant 2.1)
- 📋 Pyde Improvement Proposals (PIPs) — protocol-affecting changes go here
These apply org-wide. Individual repos may publish more specific
versions where the domain warrants it (e.g., a crate-specific security
policy on pyde-crypto once the audit kicks off).
- 📧 Email:
info@pyde.network - 🌐 Website: https://pyde.network
- 🔐 Security disclosures:
security@pyde.network(see SECURITY.md)
(Additional channels will be announced as the project moves from pre-mainnet engineering toward public testnet.)
Code is licensed under Apache-2.0 (per-repo LICENSE files).
The book is licensed under CC BY-SA 4.0.
Substantive protocol changes go through a PIP (Pyde Improvement
Proposal) — see the pips repo once it's published.