{[
- { label: t.nodeRewards, pct: 40, color: GREEN },
- { label: t.founder, pct: 20, color: GOLD },
- { label: t.community, pct: 20, color: CYAN },
- { label: t.treasury, pct: 10, color: PURPLE },
- { label: t.liquidity, pct: 10, color: '#ff6b6b' },
+ { label: t.minedByWork, pct: 100, color: GREEN },
+ { label: t.premine, pct: 0, color: GOLD },
+ { label: t.sale, pct: 0, color: '#ff6b6b' },
].map(a => (
diff --git a/apps/website/src/data/blog/bodies/tri-mined-not-sold.ts b/apps/website/src/data/blog/bodies/tri-mined-not-sold.ts
index 258b1e895a..173aac0fa0 100644
--- a/apps/website/src/data/blog/bodies/tri-mined-not-sold.ts
+++ b/apps/website/src/data/blog/bodies/tri-mined-not-sold.ts
@@ -46,26 +46,6 @@ export const body: Block[] = [
kind: 'p',
text: 'The two chain minters — a TON jetton contract and a Solana program — must reproduce that oracle exactly. The Solana side is verified host-side with cargo test: the attestation digest is byte-identical across Zig, Python and Rust (one golden vector, 9ce2cee5…), and the Ed25519 instruction parser and quorum count behave as the oracle does. A divergence is a bug in the contract, not the oracle.',
},
- {
- kind: 'h',
- text: '10 Billion Token Twin Experiment: FP vs Ternary 100M',
- },
- {
- kind: 'p',
- text: 'To test whether the quality gap between full-precision and ternary weights narrows with dataset scale, we completed a controlled twin experiment: two identical 100M parameter models (pair_fp in FP16 and pair_tern in BitNet b1.58 ternary), trained on the exact same 10.0 billion tokens of code.',
- },
- {
- kind: 'table',
-head: ['Model', 'Precision', 'Tokens', 'Val Loss', 'Val BPB'],
- rows: [
- ['pair_fp', 'FP16 (16-bit)', '10.0B', '1.1354', '0.4217'],
- ['pair_tern', 'Ternary {-1,0,1}', '10.0B', '1.2718', '0.4722'],
- ],
- },
- {
- kind: 'p',
- text: 'The ternary model achieves 0.4722 bits/byte on held-out code streams — a tight +0.0505 bpb gap (+12.0% loss) against FP16 at full 10B token scale, while running purely on additions rather than matrix multiplications.',
- },
{
kind: 'h',
text: 'What this does not establish',
@@ -122,26 +102,6 @@ export const ruBody: Block[] = [
kind: 'p',
text: 'Два контракта — jetton на TON и программа на Solana — должны воспроизводить этот оракул точно. Solana-сторона проверена на хосте через cargo test: дайджест аттестации бит-в-бит совпадает у Zig, Python и Rust (один золотой вектор, 9ce2cee5…), а разбор ed25519-инструкции и подсчёт кворума ведут себя как оракул. Расхождение — это баг контракта, а не оракула.',
},
- {
- kind: 'h',
- text: 'Близнецовый эксперимент на 10 млрд токенов: FP vs Тернарная 100M',
- },
- {
- kind: 'p',
- text: 'Чтобы проверить, сужается ли разрыв в качестве между полной точностью и тернарными весами с объёмом данных, мы завершили контрольный эксперимент: две идентичные модели на 100 млн параметров (pair_fp в FP16 и pair_tern в тернарном BitNet b1.58), обученные на одних и тех же 10.0 миллиардах токенов кода.',
- },
- {
- kind: 'table',
- head: ['Модель', 'Точность весов', 'Токены', 'Val Loss', 'Val BPB'],
- rows: [
- ['pair_fp', 'FP16 (16 бит)', '10.0 млрд', '1.1354', '0.4217'],
- ['pair_tern', 'Тернарная {-1,0,1}', '10.0 млрд', '1.2718', '0.4722'],
- ],
- },
- {
- kind: 'p',
- text: 'Тернарная модель достигает 0.4722 бит на байт на отложенном коде — плотный разрыв всего в +0.0505 bpb (+12.0% по лоссу) против FP16 на полном масштабе 10 млрд токенов, работая исключительно на сложениях вместо умножений.',
- },
{
kind: 'h',
text: 'Что это не устанавливает',
diff --git a/deploy/contracts/README.md b/deploy/contracts/README.md
index 4d3585867f..86c35c22a1 100644
--- a/deploy/contracts/README.md
+++ b/deploy/contracts/README.md
@@ -6,6 +6,25 @@ ERC20 + Permit + Vesting token for the Trinity ternary AI network.
**Sacred Formula:** phi^2 + 1/phi^2 = 3 (Trinity Identity)
+> **SUPERSEDED -- DO NOT DEPLOY.** On 2026-09-24 this contract and its
+> tokenomics were superseded by the mint-on-acceptance design in
+> [`gHashTag/trinity-fpga`](https://github.com/gHashTag/trinity-fpga) @ `d7e9718e9`:
+> `specs/trinet/mint_on_acceptance.t27`, `contracts/README.md`,
+> `docs/docs/depin/tokenomics.md`, `docs/docs/depin/principles.md`.
+> There, 100% of TRI is mined by accepted work: zero pre-mine, no allocation,
+> no sale; cap 3^21 = 10,460,353,203 TRI; chains TON and Solana; trust model
+> V1 M-of-N attestor quorum, then V2 optimistic challenge, then V3 zk receipt.
+> Nothing is deployed.
+>
+> The pre-mine and vesting allocation below contradict "100% mined". The
+> Sepolia deployment `0xef368e29FA3aB2eaf02BccD05438ED3bafE9f469`
+> (`deployment-sepolia.json`) is a historical testnet artefact: all five
+> allocation addresses equal the deployer, so the four `_setupVesting` calls
+> overwrote one another and only the last (treasury, 10%) survived. At most
+> 20% of the cap can ever exist on it; 80% can never be minted. See the
+> header of `src/TrinityToken.sol` for the derivation and the on-chain check.
+> This directory is kept as history only.
+
---
## Prerequisites
@@ -156,7 +175,7 @@ forge script script/Deploy.s.sol:DeployTrinityToken \
---
-## Tokenomics
+## Tokenomics (superseded, see the note at the top)
| Allocation | Share | Amount (TRI) | Vesting | Cliff |
|------------|-------|-------------|---------|-------|
diff --git a/deploy/contracts/src/TrinityToken.sol b/deploy/contracts/src/TrinityToken.sol
index 2a967a3f50..68a19d9dae 100644
--- a/deploy/contracts/src/TrinityToken.sol
+++ b/deploy/contracts/src/TrinityToken.sol
@@ -1,4 +1,40 @@
// SPDX-License-Identifier: MIT
+//
+// ============================================================================
+// SUPERSEDED -- DO NOT DEPLOY.
+//
+// Superseded on 2026-09-24 by the mint-on-acceptance design in
+// gHashTag/trinity-fpga @ d7e9718e9:
+// specs/trinet/mint_on_acceptance.t27 (protocol of record)
+// contracts/README.md (reference minters, TON + Solana)
+// docs/docs/depin/tokenomics.md (100% mined, zero pre-mine)
+// docs/docs/depin/principles.md (owner decision, 2026-09-24)
+// Current design: 100% of TRI is mined by accepted work; genesis supply is
+// zero; no pre-mine, no allocation, no sale; cap 3^21 = 10,460,353,203 TRI;
+// chains TON and Solana; trust model V1 M-of-N attestor quorum -> V2
+// optimistic challenge -> V3 zk receipt. Nothing is deployed.
+//
+// Why this contract is wrong under that design:
+// * It pre-mines (10% liquidity at construction) and reserves 80% more as
+// founder / node / community / treasury vesting allocations. That
+// contradicts "100% mined by accepted work".
+//
+// The Sepolia deployment 0xef368e29FA3aB2eaf02BccD05438ED3bafE9f469
+// (deployment-sepolia.json, 2026-02-16) is a historical testnet artefact.
+// In it all five allocation addresses equal the deployer
+// (0xb7183E9D4176F3Fa6F56190d677626a9c87a9f87), because script/Deploy.s.sol
+// and scripts/deploy.js fall back to the deployer when an *_ADDRESS env var
+// is unset. _setupVesting() writes per-beneficiary mappings, so each of the
+// four calls in the constructor overwrote the previous one for that single
+// address; only the last (treasury: 10%, 60 months, 6-month cliff)
+// survives. The maximum that can ever exist on that deployment is therefore
+// 10% liquidity + 10% vesting = 20% of the cap; 80% (founder 20 + node
+// rewards 40 + community 20) can never be minted. Checked on-chain via
+// eth_call on 2026-10-01: totalSupply() = totalVested(deployer) =
+// 1,046,035,320.3 TRI (10% each), vestingDuration(deployer) = 155,520,000 s.
+//
+// Kept as history. Do not change the code below; do not deploy it.
+// ============================================================================
pragma solidity ^0.8.20;
import "@openzeppelin/contracts/token/ERC20/ERC20.sol";
diff --git a/deploy/trinity-nexus/docs/research/3M_STRATEGY.md b/deploy/trinity-nexus/docs/research/3M_STRATEGY.md
index 585b03b0ae..673af92dce 100644
--- a/deploy/trinity-nexus/docs/research/3M_STRATEGY.md
+++ b/deploy/trinity-nexus/docs/research/3M_STRATEGY.md
@@ -1,3 +1,5 @@
+> **SUPERSEDED (2026-09-24).** $TRI has no sale, no SAFT and no pre-allocation: 100% is mined by accepted work. See [`SUPERSEDED.md`](SUPERSEDED.md) and gHashTag/trinity-fpga@d7e9718e9 `docs/docs/depin/principles.md`. Kept as history only.
+
# Trinity $3M Strategy (6 Months)
**φ² + 1/φ² = 3**
diff --git a/deploy/trinity-nexus/docs/research/BUSINESS_MODEL.md b/deploy/trinity-nexus/docs/research/BUSINESS_MODEL.md
index 275ac1d9e6..30255b72f4 100644
--- a/deploy/trinity-nexus/docs/research/BUSINESS_MODEL.md
+++ b/deploy/trinity-nexus/docs/research/BUSINESS_MODEL.md
@@ -1,3 +1,5 @@
+> **SUPERSEDED (2026-09-24).** $TRI has no sale, no SAFT and no pre-allocation: 100% is mined by accepted work. See [`SUPERSEDED.md`](SUPERSEDED.md) and gHashTag/trinity-fpga@d7e9718e9 `docs/docs/depin/principles.md`. Kept as history only.
+
# Trinity Network - Business Model
## Executive Summary
diff --git a/deploy/trinity-nexus/docs/research/INVESTOR_OUTREACH.md b/deploy/trinity-nexus/docs/research/INVESTOR_OUTREACH.md
index 15a493f380..99f9fa2bba 100644
--- a/deploy/trinity-nexus/docs/research/INVESTOR_OUTREACH.md
+++ b/deploy/trinity-nexus/docs/research/INVESTOR_OUTREACH.md
@@ -1,3 +1,5 @@
+> **SUPERSEDED (2026-09-24).** $TRI has no sale, no SAFT and no pre-allocation: 100% is mined by accepted work. See [`SUPERSEDED.md`](SUPERSEDED.md) and gHashTag/trinity-fpga@d7e9718e9 `docs/docs/depin/principles.md`. Kept as history only.
+
# Investor Outreach Templates
**$TRI Seed Round: $3M for 1%**
diff --git a/deploy/trinity-nexus/docs/research/SAFT_AGREEMENT.md b/deploy/trinity-nexus/docs/research/SAFT_AGREEMENT.md
index 8359314e76..26ba9cd56e 100644
--- a/deploy/trinity-nexus/docs/research/SAFT_AGREEMENT.md
+++ b/deploy/trinity-nexus/docs/research/SAFT_AGREEMENT.md
@@ -1,3 +1,5 @@
+> **SUPERSEDED (2026-09-24).** $TRI has no sale, no SAFT and no pre-allocation: 100% is mined by accepted work. See [`SUPERSEDED.md`](SUPERSEDED.md) and gHashTag/trinity-fpga@d7e9718e9 `docs/docs/depin/principles.md`. Kept as history only.
+
# SIMPLE AGREEMENT FOR FUTURE TOKENS (SAFT)
**$TRI Token — Trinity Network**
diff --git a/deploy/trinity-nexus/docs/research/SALE_EXECUTION_REPORT.md b/deploy/trinity-nexus/docs/research/SALE_EXECUTION_REPORT.md
index 95e8d7a06c..727b702a55 100644
--- a/deploy/trinity-nexus/docs/research/SALE_EXECUTION_REPORT.md
+++ b/deploy/trinity-nexus/docs/research/SALE_EXECUTION_REPORT.md
@@ -1,3 +1,5 @@
+> **SUPERSEDED (2026-09-24).** $TRI has no sale, no SAFT and no pre-allocation: 100% is mined by accepted work. See [`SUPERSEDED.md`](SUPERSEDED.md) and gHashTag/trinity-fpga@d7e9718e9 `docs/docs/depin/principles.md`. Kept as history only.
+
# $TRI Seed Round Execution Report
**Date:** 03 February 2026
diff --git a/deploy/trinity-nexus/docs/research/SALE_PREP_REPORT.md b/deploy/trinity-nexus/docs/research/SALE_PREP_REPORT.md
index f4a7550e16..f452de2c80 100644
--- a/deploy/trinity-nexus/docs/research/SALE_PREP_REPORT.md
+++ b/deploy/trinity-nexus/docs/research/SALE_PREP_REPORT.md
@@ -1,3 +1,5 @@
+> **SUPERSEDED (2026-09-24).** $TRI has no sale, no SAFT and no pre-allocation: 100% is mined by accepted work. See [`SUPERSEDED.md`](SUPERSEDED.md) and gHashTag/trinity-fpga@d7e9718e9 `docs/docs/depin/principles.md`. Kept as history only.
+
# $TRI Token Sale Preparation Report
**Date:** 03 February 2026
diff --git a/deploy/trinity-nexus/docs/research/SUPERSEDED.md b/deploy/trinity-nexus/docs/research/SUPERSEDED.md
new file mode 100644
index 0000000000..72cc4da1e1
--- /dev/null
+++ b/deploy/trinity-nexus/docs/research/SUPERSEDED.md
@@ -0,0 +1,37 @@
+# SUPERSEDED: token sale, SAFT and pre-allocation documents
+
+**Status: superseded on 2026-09-24. Kept as history. Do not use, sign or send.**
+
+The $TRI design of record is in
+[`gHashTag/trinity-fpga`](https://github.com/gHashTag/trinity-fpga) at commit
+`d7e9718e9`:
+
+- `docs/docs/depin/principles.md` -- owner decision of 2026-09-24: "100% of it
+ is mined by accepted work -- there is no pre-mine and no sale."
+- `docs/docs/depin/tokenomics.md` -- "There is no allocation." Founder/team 0%,
+ treasury / pre-sale / liquidity 0%.
+- `specs/trinet/mint_on_acceptance.t27` -- `GENESIS_MINTED 0 // no pre-mine, no
+ allocation, no sale`; cap 3^21 = 10,460,353,203 TRI; chains TON and Solana;
+ trust model V1 M-of-N attestor quorum, then V2 optimistic challenge, then V3
+ zk receipt.
+- `contracts/README.md` -- reference minters only; nothing is deployed.
+
+A token sale or a SAFT (a promise of future tokens in exchange for money)
+directly contradicts "no sale", and every pre-allocation table (founder, team,
+treasury, public sale, liquidity) contradicts "100% mined by accepted work".
+
+## Files in this directory that this note covers
+
+| File | What it proposes |
+|---|---|
+| `SAFT_AGREEMENT.md` | a Simple Agreement for Future Tokens for $TRI |
+| `SALE_PREP_REPORT.md` | preparation of a $TRI token sale |
+| `SALE_EXECUTION_REPORT.md` | execution plan for a seed round using the SAFT |
+| `TOKENOMICS.md` | a pre-allocated $TRI distribution |
+| `3M_STRATEGY.md` | a $1M token sale channel and public-sale allocation |
+| `BUSINESS_MODEL.md` | a 15% public sale / liquidity allocation |
+| `INVESTOR_OUTREACH.md` | SAFT terms in investor messaging |
+
+Any statement in these files about selling, pre-allocating or vesting $TRI is
+not the current design. Equity-financing material for the company is a separate
+matter and is not addressed by this note.
diff --git a/deploy/trinity-nexus/docs/research/TOKENOMICS.md b/deploy/trinity-nexus/docs/research/TOKENOMICS.md
index 88bb431073..59c3f8f8ae 100644
--- a/deploy/trinity-nexus/docs/research/TOKENOMICS.md
+++ b/deploy/trinity-nexus/docs/research/TOKENOMICS.md
@@ -1,3 +1,5 @@
+> **SUPERSEDED (2026-09-24).** $TRI has no sale, no SAFT and no pre-allocation: 100% is mined by accepted work. See [`SUPERSEDED.md`](SUPERSEDED.md) and gHashTag/trinity-fpga@d7e9718e9 `docs/docs/depin/principles.md`. Kept as history only.
+
# $TRI Token - Tokenomics Specification
## Overview
diff --git a/docs/docs/depin/tokenomics.md b/docs/docs/depin/tokenomics.md
index cc2c79c06c..669531c90c 100644
--- a/docs/docs/depin/tokenomics.md
+++ b/docs/docs/depin/tokenomics.md
@@ -6,6 +6,16 @@ description: 'Supply, vesting, staking — everything about $TRI'
# $TRI Token Economics
+:::danger[Superseded on 2026-09-24]
+The allocation, vesting and Sepolia deployment on this page are no longer the
+$TRI design. The design of record is in
+[gHashTag/trinity-fpga](https://github.com/gHashTag/trinity-fpga/blob/main/docs/docs/depin/tokenomics.md)
+(commit `d7e9718e9`, with `specs/trinet/mint_on_acceptance.t27`): 100% of $TRI is
+mined by accepted work, zero pre-mine, no allocation, no sale; cap
+3^21 = 10,460,353,203 TRI; chains TON and Solana; nothing is deployed. This page
+is kept as history.
+:::
+
$TRI is the native token of the Trinity DePIN network. It rewards node operators, governs protocol parameters, and serves as the unit of account for all on-chain operations.
## Total Supply
diff --git a/specs/tri27/reward_distribution.tri b/specs/tri27/reward_distribution.tri
index 23972e478b..e735a6a74f 100644
--- a/specs/tri27/reward_distribution.tri
+++ b/specs/tri27/reward_distribution.tri
@@ -1,3 +1,12 @@
+# SUPERSEDED (2026-09-24) -- not the TRI token-economics rule. Kept as history.
+# Source of truth: gHashTag/trinity-fpga @ d7e9718e9,
+# specs/trinet/mint_on_acceptance.t27 and docs/docs/depin/tokenomics.md:
+# 100% of TRI is mined by accepted work; genesis supply 0; no pre-mine,
+# no allocation, no sale; cap 3^21 = 10,460,353,203 TRI; TON + Solana;
+# nothing deployed.
+# Contradiction here: emission split 60% stakers / 30% treasury / 10% team
+# (from #547, d3d6bc7e7); the current design has no treasury or team share.
+#
# TRI-27 Reward Distribution — Source of Truth
# Issue: #444
# phi^2 + 1/phi^2 = 3 | TRINITY
diff --git a/specs/tri27/staking.tri b/specs/tri27/staking.tri
index 855755aa42..f4444be43b 100644
--- a/specs/tri27/staking.tri
+++ b/specs/tri27/staking.tri
@@ -1,3 +1,12 @@
+# SUPERSEDED (2026-09-24) -- not the TRI token-economics rule. Kept as history.
+# Source of truth: gHashTag/trinity-fpga @ d7e9718e9,
+# specs/trinet/mint_on_acceptance.t27 and docs/docs/depin/tokenomics.md:
+# 100% of TRI is mined by accepted work; genesis supply 0; no pre-mine,
+# no allocation, no sale; cap 3^21 = 10,460,353,203 TRI; TON + Solana;
+# nothing deployed.
+# Contradiction here: REWARD_EMISSION_RATE mints per second of time,
+# not per accepted piece of work.
+#
# TRI-27 Staking Module — Source of Truth
# Issue: #443
# phi^2 + 1/phi^2 = 3 | TRINITY
diff --git a/specs/tri27/token_types.tri b/specs/tri27/token_types.tri
index ce618a24a2..f51af45dc3 100644
--- a/specs/tri27/token_types.tri
+++ b/specs/tri27/token_types.tri
@@ -1,3 +1,11 @@
+# SUPERSEDED (2026-09-24) -- not the TRI token-economics rule. Kept as history.
+# Source of truth: gHashTag/trinity-fpga @ d7e9718e9,
+# specs/trinet/mint_on_acceptance.t27 and docs/docs/depin/tokenomics.md:
+# 100% of TRI is mined by accepted work; genesis supply 0; no pre-mine,
+# no allocation, no sale; cap 3^21 = 10,460,353,203 TRI; TON + Solana;
+# nothing deployed.
+# Contradiction here: MAX_SUPPLY = 270M tokens (the cap is 3^21).
+#
# TRI-27 Token Types — Source of Truth
# Issue: #441
# phi^2 + 1/phi^2 = 3 | TRINITY