Tessera keeps the Bitcoin Core design — UTXO ledger, Script, full-node validation,
P2P network, built-in wallet — and replaces the cryptography that a quantum computer
breaks. There is no elliptic-curve code in the tree at all:
libsecp256k1 is
not present.
Why nowIn March 2026 a joint whitepaper from Google Quantum AI, the Ethereum Foundation and
Stanford put a number on the threat: the 256-bit elliptic-curve discrete logarithm —
the problem every Bitcoin signature rests on — falls in minutes to fewer than half a
million physical qubits.
arXiv:2603.28846NIST has separately scheduled ECDSA for deprecation after 2030 and disallowance after
2035 (NIST IR 8547).
This is not a claim that Bitcoin is doomed tomorrow. It is a claim that a chain being
started
today has no reason to be built on a primitive with a published expiry
date. The genesis coinbase carries the first of those two findings rather than a slogan:
Google Quantum AI 30/Mar/2026 ECDLP-256 in minutes on under half a million physical qubits
Specification| Signatures | ML-DSA-44 — FIPS 204, vendored from PQClean |
| Key exchange | ML-KEM-768 — FIPS 203, in the v2 P2P transport |
| Hashing | SHA3-256 — FIPS 202, for block and transaction ids and Merkle trees |
| Proof of work | SHA3-256d over the standard 80-byte header |
| Block time | 10 minutes |
| Retarget | Bitcoin's 2016-block window, unchanged |
| Subsidy | 50 TSR, halving every 210 000 blocks |
| Ticker | TSR |
| Addresses | bech32 tsr1…, legacy prefix T |
| P2P / RPC | 11333 / 11332 |
Everything outside the cryptography is deliberately unchanged. A post-quantum chain is
already enough novelty for one project.
LaunchNo premine. No ICO. No dev fund. No masternodes.The genesis coinbase output is
OP_RETURN — provably unspendable, so the 50 TSR
of block 0 do not exist and cannot be claimed by anyone, including me.
The chain starts at
difficulty 1 — the same target Bitcoin's genesis used. That is
deliberate: an ordinary CPU finds a block in a few minutes until the difficulty finds its
level. Nobody needed a rig to be ready on day one.
Genesis: 033d419a2fc78583172e44f3f3681f1c6f83cf78e3e9becb06a255e1290f4a6e
nTime: 1786492800 (12 Aug 2026, 00:00 UTC)
nBits: 0x1d00ffff
DownloadsNode, CLI, Qt GUI wallet, CPU miner and the offline tools. Nothing to install — unpack
and run.
Windows x86-64https://gettessera.org/dl/tessera-0.1.0-win64.zip5daa5b35df79de73244d8a9364c8cabec7986c991bd63e635cb061f8382dabff
(needs glibc 2.38+ — Ubuntu 23.10, Debian 13, Fedora 39 and later)
https://gettessera.org/dl/tessera-0.1.0-x86_64-linux-gnu.tar.gz6d2b9dd128ae01d12714b8720bf50c1f879d5bf931e1aa9f1792b179454c3ed7
The same checksums are committed to the repository:
https://github.com/gettessera-project/tessera/blob/main/SHA256SUMSCompare the two. A checksum served from the same host as the file it describes proves
nothing — whoever could replace one could replace the other. Two independent sources is
the point.
Miningtessera-miner fetches work over JSON-RPC, grinds SHA3-256d on every core and
submits the block. It is a plain CPU miner. No GPU code ships in this tree — anyone who
wants one is free to write it, and the protocol does nothing to stop them.
./tesserad -daemon
./tessera-cli getnewaddress
./tessera-miner -address=<your address>
./tessera-miner -benchmark # measure this machine first
Roughly 2 MH/s per core on a current desktop; at difficulty 1 that is a few minutes per
block on eight cores.
The honest partA post-quantum signature is not free, and pretending otherwise would be the first lie in
a project whose whole point is not needing one.
An ML-DSA-44 input is
3732 bytes against 104 for ECDSA. A transaction is roughly
17× the size of a Bitcoin transaction, and at equal network congestion a payment
costs proportionally more in fees. That is the price of the guarantee, stated up front.
What is not done yet:
- No block explorer
- No exchange listings, and none sought yet
- Linux binaries need a recent glibc; a broadly compatible build needs an older build environment
- Translations were inherited from Bitcoin Core and are not reviewed by native speakers
Consensus rules may still change before the network has real weight behind it. After that
they will not.
LinksWebsite:
https://gettessera.orgSource:
https://github.com/gettessera-project/tesseraDiscord:
https://discord.gg/trFG8BD3wSecurity: report privately via GitHub Security Advisories, never in a public issue
MIT licensed. Built on the Bitcoin Core architecture, with respect to its authors.
