Get basic research reports
curl --request GET \
--url https://api.messari.io/research-basic/v1/basic/reports \
--header 'X-Messari-API-Key: <api-key>'import requests
url = "https://api.messari.io/research-basic/v1/basic/reports"
headers = {"X-Messari-API-Key": "<api-key>"}
response = requests.get(url, headers=headers)
print(response.text)const options = {method: 'GET', headers: {'X-Messari-API-Key': '<api-key>'}};
fetch('https://api.messari.io/research-basic/v1/basic/reports', options)
.then(res => res.json())
.then(res => console.log(res))
.catch(err => console.error(err));<?php
$curl = curl_init();
curl_setopt_array($curl, [
CURLOPT_URL => "https://api.messari.io/research-basic/v1/basic/reports",
CURLOPT_RETURNTRANSFER => true,
CURLOPT_ENCODING => "",
CURLOPT_MAXREDIRS => 10,
CURLOPT_TIMEOUT => 30,
CURLOPT_HTTP_VERSION => CURL_HTTP_VERSION_1_1,
CURLOPT_CUSTOMREQUEST => "GET",
CURLOPT_HTTPHEADER => [
"X-Messari-API-Key: <api-key>"
],
]);
$response = curl_exec($curl);
$err = curl_error($curl);
curl_close($curl);
if ($err) {
echo "cURL Error #:" . $err;
} else {
echo $response;
}package main
import (
"fmt"
"net/http"
"io"
)
func main() {
url := "https://api.messari.io/research-basic/v1/basic/reports"
req, _ := http.NewRequest("GET", url, nil)
req.Header.Add("X-Messari-API-Key", "<api-key>")
res, _ := http.DefaultClient.Do(req)
defer res.Body.Close()
body, _ := io.ReadAll(res.Body)
fmt.Println(string(body))
}HttpResponse<String> response = Unirest.get("https://api.messari.io/research-basic/v1/basic/reports")
.header("X-Messari-API-Key", "<api-key>")
.asString();require 'uri'
require 'net/http'
url = URI("https://api.messari.io/research-basic/v1/basic/reports")
http = Net::HTTP.new(url.host, url.port)
http.use_ssl = true
request = Net::HTTP::Get.new(url)
request["X-Messari-API-Key"] = '<api-key>'
response = http.request(request)
puts response.read_body{
"error": null,
"data": [
{
"id": "16d6619b-9416-46cc-8fd2-13fdcd9bae2b",
"createdAt": "2026-04-06T23:37:35Z",
"updatedAt": "2026-04-07T16:00:58Z",
"assetIds": [
"2023433a-23f4-4901-822d-a537b0c71676"
],
"assets": [
{
"id": "2023433a-23f4-4901-822d-a537b0c71676",
"name": "Toncoin",
"symbol": "TON",
"slug": "the-open-network"
}
],
"authors": [
{
"id": "872be9a6-e5d6-47c4-ac6d-95bb1720adf9",
"name": "Whynonah",
"image": "https://cdn.sanity.io/images/2bt0j8lu/production/8ac8076d5648490dd40511ab3b81c602a03be811-681x793.jpg?w=100",
"linkedinUrl": ""
},
{
"id": "2b10859c-dd75-472e-afde-1d1df4d2f576",
"name": "Jonny Kreiser",
"image": "https://cdn.sanity.io/images/2bt0j8lu/production/5f0238a98921e3b37e6b3034ebe88c6561fd393c-400x400.jpg?w=100",
"linkedinUrl": ""
}
],
"content": "## **Key Insights**\n\n* **Toncoin (TON) functions as the network’s core execution asset (i.e., paying gas, supporting DeFi liquidity, securing Proof-of-Stake consensus, and acting as a settlement layer)** and as the foundation of Telegram's in-app economy.\n* **2025 activity shows TON shifting from early-year viral surges into a steadier baseline** (~100K–150K daily active users and ~1.5–2.5 million daily transactions).\n* **TON’s ecosystem** is evolving into a Telegram-native financial stack, with stablecoins, yield products, and RWAs increasingly embedded in wallets and mini-apps; **highlighted by xStocks, Ethena, Tether, Affluent, and liquidity routing via STON.fi + Omniston.**\n* **TON's 2026 roadmap pivots from infrastructure refactoring to performance and developer accessibility**. Catchain 2.0 targets sub-second finality, the Rust Node reimplements the validator stack, and a unified developer layer that includes developer kits for smart contracts, apps, wallets and payments.\n* **TON is emerging as a native infrastructure layer for AI** inside Telegram, with Cocoon providing decentralized compute, AgentKit connecting autonomous agents to onchain actions, and vibe-coding workflows enabling builders to quickly generate and share working prototypes.\n\n## **Introduction**\n\nConsumer crypto adoption typically breaks down into two bottlenecks: distribution and cost/latency. Most blockchains rely on standalone wallets and browser-based dApps, creating multi-step onboarding funnels that deter mainstream users. Even when infrastructure works, blockchain interactions often feel slow, expensive, or fragmented compared to Web2 applications; particularly in consumer use cases like payments, gaming, and social applications, where users expect instant feedback and low friction.\n\nTON’s thesis is that mass adoption requires both scalable infrastructure and native distribution. At the base layer, TON is built as a dynamically sharded Proof-of-Stake (PoS) network with asynchronous smart contracts, allowing parallel execution and horizontal scalability. Rather than relying solely on high single-chain throughput, TON’s architecture is designed to maintain stable performance under load. Complementing the Layer-1 are native protocol services, including (i) [TON DNS](https://dns.ton.org/) (human-readable “.ton” names), (ii) [TON Storage](https://docs.ton.org/foundations/services#ton-storage) (decentralized file storage), (iii) [TON Payments](https://docs.ton.org/foundations/services#ton-payments) (payment channels), (iv) [TON Proxy](https://docs.ton.org/foundations/services#ton-proxy) (censorship-resistant routing), (v) [TON Sites ](https://ton.org/en/ton-sites)(decentralized websites), and (vi) [Tolk](https://docs.ton.org/languages/tolk/overview#tolk-language) (smart contract language), and (vii) [AppKit](https://docs.ton.org/ecosystem/appkit/overview) (an all-in-one SDK for building Telegram Mini Apps with TON), which together form a vertically integrated blockchain stack with easy developer tooling.\n\nWhat differentiates TON is its distribution and tight integration with [Telegram](https://telegram.org/). [**TON Wallet**](https://wallet.tg/ton), natively embedded in the messenger, allows users to transact onchain without leaving the app. [**TON Connect**](https://docs.ton.org/ecosystem/ton-connect/overview), the wallet-connection protocol for **Telegram Mini Apps (TMAs)**, opens this to any compatible third-party wallet, though TON Wallet remains unique as Telegram's built-in wallet. Combined with in-chat app distribution, the result is onchain actions that feel like normal in-app behaviour, collapsing the traditional crypto UX stack into a single consumer environment.\n\n[Website](https://ton.org/en) / [X (Twitter)](https://x.com/ton_blockchain) / [LinkedIn](https://www.linkedin.com/company/ton-blockchain)\n\n\n\n## **Background**\n\nTON was founded in 2018 as the “Telegram Open Network” by Telegram co-founders [Pavel Durov ](https://x.com/durov?lang=en)and [Nikolai Durov](https://x.com/Kolja_Durov?lang=en), and raised $1.7 billion across two private token sales in February–March 2018 to fund development. In October 2019, [the](https://aurum.law/newsroom/telegram-ton-1-7-bill-raise-sec-decentralization-the-legal-tale-and-insights) U.S. Securities and Exchange Commission (SEC) filed an emergency action against Telegram alleging an unregistered token offering, and Telegram ultimately ceased active involvement in May 2020. Development continued through the community-led “Newton” effort initiated by [Anatoliy Makosov](https://x.com/anatoly_makosov) and Kirill Emelyanenko, and in May 2021, the community voted to formalize governance under the TON Foundation and promote the V2 testnet into TON Mainnet.\n\nFollowing the community relaunch, TON attracted additional strategic and private investment. Since 2022, TON has [completed](https://messari.io/project/the-open-network/fundraising/funding) eight additional funding rounds, with at least three publicly disclosed raises totaling $50 million ($10 million from [DWF Labs](https://messari.io/organization/dwf-labs) in 2022, $30 million in a private sale backed by Foresight Ventures and Bitget in 2024, and $10 million in a strategic investment from [Gate](https://messari.io/organization/gate-io) in 2024).\n\nSince 2022, TON has attracted significant institutional backing. In [March 2025](https://www.tradingview.com/news/cointelegraph:d27a07a06094b:0-venture-capital-firms-invest-400m-in-ton-blockchain/), the TON Foundation disclosed that a group of investors, including Sequoia Capital, Pantera Capital, and Ribbit Capital, collectively purchased over $400 million in Toncoin. Combined with earlier rounds and subsequent investments from Coinbase Ventures, Pantera Capital, and others, publicly confirmed capital inflows exceed $550 million. Later in [August 2025](https://cryptoslate.com/verb-secures-558m-to-become-first-ton-treasury-vehicle-plans-ton-strategy-rebrand/), two publicly traded Toncoin treasury vehicles, [TON Strategy Co.](https://cryptoslate.com/verb-secures-558m-to-become-first-ton-treasury-vehicle-plans-ton-strategy-rebrand/) and [AlphaTON Capital](https://decrypt.co/news-explorer?pinned=1150165&title=portage-biotech-rebrands-as-alphaton-capital-corp-invests-100m-in-ton-digital-asset-treasury), have also launched, modeled on MicroStrategy's BTC accumulation strategy.\n\nIn [March 2025](https://www.tradingview.com/news/cointelegraph:d27a07a06094b:0-venture-capital-firms-invest-400m-in-ton-blockchain/), TON Foundation disclosed that a group of investors - including Sequoia Capital, Ribbit Capital, Benchmark, Draper Associates, Kingsway Capital, Vy Capital, Libertus Capital, CoinFund, SkyBridge, Hypersphere, and Karatage - collectively purchased and held $400 million in Toncoin. Pantera Capital, which has called Toncoin its largest single investment to date, reportedly deployed over $100 million. In [August 2025](https://www.theblock.co/post/366414/coinbase-ventures-buys-ton-tokens-directly-from-telegram-says-ton-foundation-exec), Coinbase Ventures purchased Toncoin directly from Telegram as a long-term venture position. In total, publicly confirmed capital inflows into TON exceed $550 million since 2022 across at least ten discrete investment events.\n\nA later turning point came through renewed alignment with Telegram distribution. In September 2023, the TON Foundation [announced](https://wublock.substack.com/p/ton-ecosystem-overview-deep-integration) a partnership with Telegram, aiming to integrate and promote the TON ecosystem for Telegram’s userbase. Thanks to this partnership, Telegram natively integrated a crypto wallet known as @Wallet, supported TON Sites in an in-app browser, and issued Fragment collections (usernames, collectable numbers, and gifts) on TON.\n\nA later turning point came through renewed alignment with Telegram distribution. In January 2025, TON [became](https://www.prnewswire.com/news-releases/ton-foundation-expands-partnership-with-telegram-as-ton-becomes-the-exclusive-blockchain-for-telegrams-mini-app-platform-302356251.html#:~:text=Jan%2021%2C%202025%2C%2010:,950%20million%20monthly%20active%20users.) the exclusive blockchain infrastructure for Telegram’s Mini App platform, with TON Connect positioned as the standard wallet connection method and Toncoin (TON) serving as the exclusive cryptocurrency for non-fiat payments across Telegram services (e.g., Stars, Premium, Ads). This shifted TON to a technically differentiated L1 focused on consumer distribution, with Telegram serving as its primary onboarding and application surface.\n\nToday, TON is maintained through a foundation and open-source contributors, so leadership is best understood through its public-facing executives and operating entities rather than a single corporate team. Makosov and Emelyanenko are cited as founding/core members of the TON Foundation. [Steve Yun](https://www.linkedin.com/in/steve-yun/) served as President of the TON Foundation Council and later launched the $100 million ecosystem venture fund, [TVM Ventures](https://www.linkedin.com/company/tvm-ventures/), in February 2025, while remaining on the board. [Maximilian Crown](https://www.linkedin.com/in/maxcrown/) was appointed CEO in April 2025 and later President in August 2025, with prior experience as a [MoonPay](https://www.moonpay.com/) co-founder (CFO/COO).\n\n\n\n## **Technology**\n\n\n\nTON’s architecture is commonly described as a “blockchain of blockchains.” Instead of relying on a single monolithic chain, TON is built as a hierarchical system in which a top-level chain coordinates with multiple parallel chains, which can be further subdivided. This structure is designed to enable scalability at the consumer scale while maintaining interoperability and shared security. At a high level, TON consists of three core components: (i) the masterchain, (ii) the workchains, and (iii) the shardchains, which are explained below in more detail.\n\n### **Architecture**\n\n#### **Masterchain**\n\nThe masterchain is the top-level coordinating chain. It does not process regular user transactions. Instead, it maintains critical network state, including (i) network configuration parameters, (ii) the validator set and their stakes, and (iii) references to the latest finalized blocks of all workchains and shardchains. In simple terms, the masterchain acts as the global source of truth for the TON network, ensuring consistency and finality across all parallel chains.\n\n#### **Workchains**\n\nWorkchains operate in parallel under the coordination of the masterchain. Each workchain can, in theory, define its own rules (e.g., virtual machine, token standards, or execution logic) while remaining interoperable within the TON ecosystem. The architecture theoretically supports up to 2³² workchains, though in practice, only the base workchain is currently active in most production contexts. The workchain layer enables long-term flexibility, allowing TON to support specialized environments without fragmenting security.\n\n#### **Shardchains**\n\nFrom there, each workchain can be split into shardchains, i.e., smaller partitions of the network state. Sharding allows transactions and smart contract execution to be processed in parallel across multiple shards rather than sequentially on a single chain. TON’s design theoretically supports up to 2⁶⁰ shardchains per workchain, although the network dynamically creates and merges shards as needed. The defining feature of TON’s scalability model is dynamic sharding. Rather than permanently fixing the number of shards, shardchains split automatically under high load to increase processing capacity and merge when activity declines, reducing unnecessary overhead. This allows network capacity to expand or contract in response to real-time demand. The goal is to maintain stable performance and predictable transaction times even as user activity scales to consumer levels.\n\n#### **Consensus**\n\nTON uses a Proof-of-Stake (PoS) consensus model. Validators are selected based on the amount of TON staked and participate in block production and validation. Consensus is achieved using a Byzantine Fault Tolerant (BFT) protocol called Catchain, which is specifically designed to operate efficiently in a sharded environment. Catchain enables validators to coordinate across shards while preserving security and finality guarantees.\n\n\n\nTON Core is [releasing Catchain 2.0](https://t.me/toncore/99), a consensus upgrade targeting sub-second finality to bring the onchain experience closer to Web2 responsiveness. The upgrade reduces block intervals from ~2.5s to 200–400ms, delivers roughly 2.5–5x throughput improvement, and cuts finalization lag from ~10s to ~1s.\n\nDevelopment is complete: the testnet, updated on January 23, 2026, has been running stably at ~450ms block intervals with ~1–2s finalization. Mainnet validators were updated on February 12, 2026 with the Catchain 2.0 code and accelerated network layer, though activation remains dormant pending final testnet validation.\n\n### **Validators**\n\nWithin the above architecture, there are two primary roles: (i) Validator and (ii) Nominators.\n\nValidators [secure](https://ton.org/en/validators) TON’s PoS network by staking Toncoin to participate in block production and validating transactions across shardchains. Operators must run high-performance, highly available infrastructure and stake a minimum of 300,000 TON to enter validator elections, though winning typically requires ~700,000 TON or more, depending on competition and the cycle's validator cap. Validators stake for a fixed validation term, with stake and rewards returned after the round completes.\n\nValidators [earn rewards](https://docs.ton.org/v3/documentation/nodes/validation/staking-incentives) from (i) transaction fee surpluses (users attach small extra Toncoin amounts as validator incentives) and (ii) newly issued Toncoin with proportional distribution based on stake weight. Validators can be [penalized](https://docs.ton.org/v3/documentation/smart-contracts/contracts-specs/governance) in two ways: (i) idle behavior: If a validator fails to participate in block creation or transaction signing during a validation round, it may be fined; and (ii) malicious misbehavior, where any network participant can submit a complaint with cryptographic proof to the [Elector contract](https://docs.ton.org/v3/documentation/smart-contracts/contracts-specs/governance). Validators vote on the complaint, and if 66.0% of validators approve, a slashing penalty is deducted from the validator's stake. To receive rewards, a validator must successfully (i) win election into a validation cycle, and (ii) validate blocks throughout the entire cycle without being penalized.\n\n[Nominators](https://docs.ton.org/v3/documentation/smart-contracts/contracts-specs/nominator-pool), on the other hand, (i) delegate TON to validators, and (ii) share in staking rewards. They do not operate the infrastructure directly. The cited minimum delegation amount is ~10,000 TON. Nominators increase validator stake weight and help decentralize participation in consensus.\n\nBeyond direct nomination, [pooled staking](https://docs.ton.org/ecosystem/staking/overview) services such as TON Whales, Kiln, ChorusOne, and P2P aggregate smaller stakes to meet validator thresholds. Alternatively, [liquid staking](https://docs.ton.org/ecosystem/staking/overview) protocols (Tonstakers, KTON, bemo, Hipo) issue transferable receipt tokens that let holders earn staking rewards while keeping capital liquid across DeFi.\n\n### **AI**\n\nTelegram is emerging as a native interface for AI agents, with TON positioning itself as the settlement and infrastructure layer for this convergence.\n\n[Cocoon](https://cocoon.org/) (Confidential Compute Open Network) is a decentralized AI compute network built on TON, [announced](https://x.com/durov/status/1995208789600182391?lang=en) by Pavel Durov at Blockchain Life 2025 and launched on mainnet in December 2025. GPU owners contribute computing power and earn Toncoin, while user data remains encrypted throughout execution via Trusted Execution Environments (TEEs). Telegram already routes lightweight AI operations through Cocoon, including message translation and summarization, with heavier workloads such as media processing and conversational assistants planned next. With Telegram's 1B+ user base as built-in distribution and Telegram’s recent efforts to simplify the creation of [agentic products](https://telegram.org/blog/ai-editor-mighty-polls-and-more), Cocoon represents one of the most significant real-world deployments of decentralized AI compute to date.\n\nTelegram's bot infrastructure has also evolved toward AI-native use cases. In early 2026, Telegram shipped [streaming responses for bots](https://t.me/BotNews/113) (purpose-built for AI assistants) and threaded conversations for multi-topic chats. Third-party traction reinforces the direction: OpenClaw, the most popular self-hosted AI assistant framework, defaults to Telegram as its messaging layer. Additionally, a [Telegram developer](https://x.com/steipete/status/2037197024081195188) became its new maintainer. [BotFather](https://identityhub.app/blog/telegram-default-ai-interface), Telegram's built-in tool for creating and configuring bots, reached 7.3M monthly active users (MAUs) by early 2026, more than doubling from 3.5 million in January 2025. Combined with native Toncoin withdrawal support for bot earnings, these updates tighten the economic link between Telegram's bot ecosystem and the TON network.\n\nTON's deeper integration with onchain AI agents remains at an earlier stage. To accelerate development, TON Foundation is running dedicated [AI contests](https://identityhub.app/contests) to bootstrap tooling and adoption.\n\n## **Putting It All Together**\n\nThe components described above (i.e., masterchain coordination, dynamic shardchains, asynchronous smart contracts, TVM execution, cell-based data storage, and hypercube routing) operate within a broader multi-layer architecture designed for consumer-scale usage inside Telegram. Rather than functioning as a single monolithic execution engine, TON operates as an interconnected system composed of: (i) User Interface Layer, (ii) Application Layer, (iii) Execution Layer, (iv) Routing & Sharding Layer, (v) Consensus Layer, (vi) Validator Infrastructure Layer, (vii) Indexing & API Layer, (viii) Storage & Data Layer. Together, these layers allow TON to support real-time financial interactions embedded directly inside Telegram’s social graph.\n\nFor example, consider a user sending USDT to a friend inside Telegram:\n\n* The transaction begins inside the Wallet in Telegram. The user enters an amount and taps send. The experience feels identical to sending a message.\n* The wallet constructs a Jetton (TEP-74) transfer transaction. Because TON uses an asynchronous model, a message is sent to the sender’s USDT contract, which then sends a follow-up message to the recipient’s wallet contract.\n* The TON Virtual Machine executes the smart contract logic:\n * Validates balances\n * Deducts TON for gas\n * Updates Jetton balances\n * Emits internal messages\n * Each contract runs independently\n* If the sender and receiver reside on different shardchains, the message is routed using Hypercube Routing. Rather than broadcasting globally, the system calculates the shortest path between shards, minimizing latency.\n* If network load increases, shardchains may automatically split to distribute execution load. This is TON’s “infinite sharding” in action.\n* Validators on the relevant shardchain produce and validate the block. The masterchain later finalizes shard references to ensure global consistency.\n* Collators assemble transactions while validators confirm them, enabling parallelization and improved stability.\n* Toncenter APIs and indexers immediately update transaction status. Wallets can display “pending” and then a confirmed state using trace APIs and action parsing.\n* All state changes are stored in TON’s cell-based structure and packaged into a Bag-of-Cells (BoC), ensuring compact storage and verifiable hash-linked data.\n\nTo the user, this entire multichain, asynchronous, routed, validated process appears as a simple chat-based transfer.\n\n\n\n## **Toncoin (TON Token)**\n\n### **Token Functions**\n\nAs the project’s documentation outlines, Toncoin is a native coin of the TON blockchain. It serves several key functions within the network, including:\n\n* **Paying for network execution:** Toncoin is required to execute transactions (e.g., asset transfers and swaps), with fees paid in Toncoin and designed to remain low for consumer-scale usage.\n* **Securing the network:** Validators stake Toncoin to participate in PoS consensus and earn rewards, with penalties for downtime or misbehavior. Nominators can delegate Toncoin to share in validator staking rewards.\n* **Telegram in-app economy:** Telegram uses TON as its blockchain layer for ownership, transfers, and payouts. Toncoin serves as the payment rail for collectible usernames, SIM-less phone numbers, and collectible gifts - all of which are onchain assets. Creator earnings from ads and in-app purchases (for channel authors, bot developers, and mini-app owners) are withdrawn exclusively via **Fragment.com** in Toncoin. Toncoin is also the currency for purchasing ads on the **Telegram Ads** platform.\n* **Storage of blockchain data:** Toncoin is required to maintain smart contracts onchain, where validators charge storage rent proportional to the data held. For larger files, TON Storage provides a decentralized persistence layer where storage providers are paid in Toncoin and must cryptographically prove file integrity to claim rewards.\n\n### **Tokenomics**\n\n\n\nThe initial TON supply of 5 billion tokens was placed into [20](https://ton.org/en/mining) Proof-of-Work Giver smart contracts and mined permissionlessly between July 2020 and June 2022. Since the PoS transition (June 28, 2022), new TON is minted via validator block rewards (~0.5–0.7% annually). The current total supply is ~5.16 billion TON. At [$1.33](https://www.coingecko.com/en/coins/toncoin) (Mar. 26, 2026), this implies a fully diluted valuation (FDV) of ~$6.9 billion. Note: TON has no max supply cap.\n\nThe pie chart above depicts the current supply distribution of ~5.15 billion Toncoin as of March 2026, broken into 14 categories. For simplicity, these can be grouped into four macro buckets:\n\n* **Freely Circulating:** ~48% (~2.46B TON). Includes Regular Wallets, Staking (Elector), Centralized Exchanges, CEX Custodial, DeFi, Smart Contracts & Others, Uninit Wallets, TON Ecosystem Reserve, TON Foundation, and Other Labeled.\n* **TON Believers Fund:** 25% (~1.32B TON). A voluntary lockup where existing holders deposited tokens into a [Locker smart contract](https://github.com/ton-blockchain/locker-contract) (Jul–Oct 2023). Now vesting in 36 monthly installments (~36.6M TON/month) through Oct 2028. As of March 2026, 6 of 36 periods have been completed; actual claim rates have been low, with only a fraction of unlocked tokens withdrawn so far. [Onchain data.](https://tonviewer.com/UQDtFpEwcFAEcRe5mLVh2N6C0x-_hJEM7W61_JLnSF74p9dz)\n* **Frozen Inactive Miners:** 20.9% (~1.08B TON). 171 addresses that mined TON during the PoW phase but never transacted. [Frozen by community governance vote](https://blockworks.co/news/ton-governance-votes-to-freeze-20-of-its-ton-supply) (Feb 2023) for 48 months. After expiration (~Feb 2027), owners must manually activate, and many keys are likely lost. The community could vote to extend the freeze or burn these tokens before expiration, though no formal proposal has surfaced as of this writing. [Address list.](https://tontech.io/stats/#/early-miners)\n* **Telegram:** ~6% (~327M TON). Tokens held by Telegram, plus ~21M in vesting contracts deployed to team members and partners (1,440-day vesting, 360-day cliff). Sell pressure from vesting recipients has been negligible thus far.\n\n### **Token Vesting**\n\n[Unlock schedules](https://defillama.com/unlocks/ton) for the parties mentioned above include:\n\n* **IPoW Mining (Premine):** Fully mined and distributed. The original ~5B TON was mined from [20 Giver contracts](https://ton.org/en/mining) between July 6, 2020, and June 28, 2022. Mining was permissionless with a variable rate based on PoW difficulty, not a linear release schedule. No further unlocks.\n* **TON Believers Fund:** Unlocks through 36 fixed monthly releases of ~36.59 million TON/month (~2.8% of the fund). As of March 2026, 6 of 36 periods have been completed; 30 periods remain (~1.098 billion TON still to unlock through ~Oct. 2028).\n* **Frozen Inactive Miners:** The 48-month [freeze](https://blockworks.co/news/ton-governance-votes-to-freeze-20-of-its-ton-supply) on 171 addresses (~1.081 billion TON) expires around Feb. 2027. This does not mean tokens enter circulation: these wallets have never transacted, owners must manually activate them, and many keys are likely permanently lost. The TON community could also vote to extend the freeze or burn these tokens before expiration. [Address list.](https://tontech.io/stats/#/early-miners)\n* **[Vesting contracts](https://github.com/ton-blockchain/vesting-contract):** Use a 1,440-day schedule with a 360-day cliff. Most early contracts have already fully vested; remaining contracts continue vesting at ~3.45M TON/month (per CEX listing model).\n* **PoS Emission (Inflation):** Ongoing, perpetual issuance at[ ~73,000–97,000 TON/day](https://www.tonstat.com/) (1.7 TON per masterchain block + 1.0 TON per basechain block). 50% of transaction fees are [burned](https://dune.com/ton_foundation/staking).\n\n### **Governance**\n\nTON's governance spans three layers: validators, core development, and ecosystem coordination.\n\nAt the protocol layer, approximately 400 validators distributed across 40 countries, with over 450M Toncoin staked, govern upgrade decisions directly. Any change to network parameters or consensus rules must pass an onchain vote among active validators, ensuring no single entity can push through protocol changes unilaterally. According to Chainspect, TON ranks 3rd among Layer-1s by Nakamoto coefficient, placing it among the most decentralized Proof-of-Stake networks.\n\nTON Core, the network's core development arm, maintains the TON node software, ships protocol upgrades, and builds node tooling. The TON Foundation, a non-profit dedicated to the ecosystem's long-term growth, provides grants, resources, and technical support to projects building on TON.\n\nBeyond these two bodies, a growing set of independent teams contributes to infrastructure and developer tooling. TonTech, an engineering team supported by the TON Foundation, maintains core developer primitives including AppKit, AgentKit, WalletKit, and TON Connect. RSquad, a blockchain development team active in the TON ecosystem since its early days, has contributed critical infrastructure, including the Rust TON Node, TON Pay, and TON Teleport, a trustless cross-chain bridge for asset transfers between TON and external networks.\n\n## **TON Ecosystem**\n\n\n\n### **Partners and Projects**\n\nKey projects that highlight the variety of benefits the TON ecosystem and blockchain provide:\n\n* **[The Open Platform (TOP)](https://top.co/):** The largest Web3 product development company within the Telegram ecosystem, building and investing in infrastructure and consumer applications on TON. TOP's portfolio includes Wallet in Telegram, Tonkeeper, STON.fi, and Getgems, and the company reached a [$1 billion valuation](https://www.theblock.co/press-releases/361030/the-open-platform-is-first-unicorn-in-web3-ecosystem-in-telegram-at-1bn-valuation) in 2025 after raising over $70 million from Ribbit Capital, Pantera Capital, and others.\n* **[Wallet in Telegram](https://wallet.tg/ton):** A crypto wallet natively integrated into Telegram, developed by TOP, supporting both custodial and self-custodial modes dependent on region. It serves as the primary onramp for Telegram's 1B+ user base, enabling in-chat transfers, Toncoin purchases, and direct access to Telegram Mini Apps without leaving the messenger.\n* **[Tether (USDT)](https://tether.to/):** The dominant stablecoin on TON by circulating supply and the default asset for payments and most DeFi activity across the ecosystem. TON currently holds roughly [$1.28 billion in stablecoins](https://defillama.com/stablecoin/tether), with USDT serving as the primary settlement unit for wallets, DEXs, and merchant payments.\n* **[Ethena](https://ethena.fi/):** Introduced synthetic yield-bearing digital dollars (USDe and sUSDe) into TON's Telegram-native DeFi stack, expanding stablecoin use cases beyond payments into passive yield products. Eligible users holding [tsUSDe](https://blog.ton.org/ethena-ton-foundation-usde-on-ton) in a TON wallet earn boosted yields, with plans for neobanking and peer-to-peer payments powered by Ethena within Telegram.\n* **[xStocks](https://www.ton.org/en/x-stocks-are-live-on-ton-real-world-stocks-now-on-chain):** Tokenized U.S. equities launched on TON, bringing real-world stock exposure (e.g., Apple, Tesla, Microsoft) directly into TON wallets with a self-custodial UX. The platform currently supports over 60 tokenized stocks and ETFs powered by [Backed Finance](https://www.coindesk.com/business/2026/03/10/kraken-s-tokenized-stock-venue-starts-points-program-hinting-at-possible-ecosystem-token/) under Kraken's institutional framework, with plans to expand to 500+ by end of 2026.\n* **[Fragment](https://fragment.com/):** Onchain marketplace integrated into Telegram where collectible usernames, SIM-less phone numbers, and digital gifts are minted and traded as NFTs powered by Toncoin. Fragment is the primary driver of TON's [#2 ranking in NFT trading volume](https://dune.com/ton_foundation/nft) behind Ethereum, with transaction activity tied directly to Telegram's social graph.\n\nOther notable projects building on TON include [Affluent](https://www.affluent.org/), [Bidask](https://bidask.finance/), [Boinkers](https://t.me/boinker_bot), [CapsGame](https://t.me/capsgamebot), [DeDust](https://dedust.io/), [Gamee](https://t.me/gamee), [Gatto](https://t.me/gattoton_bot), [Getgems](https://getgems.io/), [MyTonWallet](https://mytonwallet.app/), [RedStone](https://redstone.finance/), [Storm Trade](https://stormtrade.io/), [Swap Coffee](https://swap.coffee/), [TeleTON](https://teletonagent.dev/), [TONCO](https://tonco.io/), [Torch Finance](https://torch.finance/), [xRocket](https://t.me/xrocket_bot), and [Zargates](https://zargates.com/).\n\nRecent partnerships and integrations:\n\n* [**Mar. 31, 2026**](https://www.tradingview.com/news/cointelegraph:11581f704094b:0-dynamic-adds-embedded-wallet-infrastructure-to-ton-for-telegram-apps/): [Dynamic](http://www.dynamic.xyz) launched embedded wallet infrastructure for TON, letting developers automatically deploy wallets inside their apps and Telegram Mini Apps.\n* **[Mar. 26, 2026](https://blockchain.news/flashnews/walletconnect-integrates-with-ton-blockchain-for-seamless-connectivity):** [WalletConnect](https://walletconnect.com/) launched production support on TON, enabling standardized wallet connections across dApps and Telegram Mini Apps.\n* **[Feb. 17, 2026](https://www.theblock.co/post/390138/ton-foundation-osl-banxa-stablecoin-payments):** [TON Foundation](https://ton.foundation/en) partnered with OSL's Banxa to expand stablecoin payment infrastructure for Asia-Pacific merchants.\n* **[Jan. 7, 2026](https://x.com/ton_blockchain/status/2008966919530090821):** TON announced Toncoin support in [Atomic Wallet](https://atomicwallet.io/), expanding TON's reach into a multichain, non-custodial wallet user base.\n* **[Dec. 24, 2025](https://x.com/ton_blockchain/status/2003764859138199985):** [Fonbnk](https://www.fonbnk.com/) + [Tether](https://tether.to/) expanded USDt on Telegram Wallet, positioning mobile-money conversion as the core bridge for Africa-focused remittances and stablecoin banking flows.\n* **[Dec. 23, 2025](https://x.com/ton_blockchain/status/2003490653645246825):** [HoudiniSwap](https://houdiniswap.com/) launched private TON payments, allowing users to request payments without exposing wallet addresses or transaction history, with inbound support across 120+ chains.\n* **[Dec. 22, 2025](https://investors.shift4.com/news-events/press-releases/detail/288/shift4-launches-global-stablecoin-settlement-platform-unlocking-faster-payments-for-merchants):** [Shift4](https://www.shift4.com/) launched a global stablecoin settlement platform, unlocking faster payments for merchants.\n* **[Aug. 5, 2025](https://x.com/ton_blockchain/status/1952723911130456184):** [Zengo](https://zengo.com/) wallet added native Toncoin support, expanding TON's presence across non-custodial mobile wallets.\n* **Upcoming:** [Chainlink CCIP](https://chain.link/cross-chain) cross-chain interoperability integration (announced 2025, launch pending). [Revolut](https://www.revolut.com/) soft-launched TON support in select regions, with broader availability now live.\n* Toncoin is also listed on major U.S. exchanges, including [Robinhood](https://thedefiant.io/news/nfts-and-web3/robinhood-adds-toncoin-to-u-s-crypto-platform-ahead-coinbase-563a1ec6) (Aug. 2025), [Gemini](https://www.gemini.com/blog/toncoin-usdton-is-now-available-on-gemini) (Sep. 2025), and [Coinbase](https://www.businesswire.com/news/home/20251118274287/en/TON-Strategy-Company-Welcomes-Coinbases-Launch-of-$TON-Trading-Across-Global-Platforms) (Nov. 2025).\n\nFor a recap of category-specific ecosystem developments, please read the following [DeFi](https://blog.ton.org/defi-on-ton), [Institutional](https://blog.ton.org/institutions-on-ton), and [NFT](https://blog.ton.org/how-nfts-evolved-on-ton) reports.\n\n### **Network Metrics**\n\n\n\nTON’s network activity in 2025 reflects a chain that has moved past the initial hype spike and is settling into a more durable, consumer-driven usage baseline. Daily active users peaked sharply in early 2025 (~600K), then normalized throughout the year, finishing Q4 2025 with ~1.0% QoQ growth and a relatively stable range of ~100K–150K.\n\n\n\nIn parallel, TON has sustained meaningful throughput, with daily transactions spiking above ~7 million during early 2025 surges, then stabilizing around ~1.5–2.5 million/day, with periodic bursts and an end-of-year lift similar to daily user data.\n\nTON's NFT market ranks [second only to Ethereum by trading volume](https://dune.com/ton_foundation/nft), driven largely by Telegram-native assets like collectible usernames, numbers, and gifts.\n\n## **Roadmap**\n\nOn Jan. 22, 2025, the TON Core team [published](https://t.me/toncore/6) its H1 2025 roadmap, centered on shipping the long-in-the-works “Accelerator” upgrade, an architecture-level refactor intended to better realize the sharded execution model described in the TON whitepaper and to keep performance stable as load scales. The roadmap prioritized (i) scaling and stability at the protocol layer, (ii) validator operability and resilience, and (iii) faster, more human-readable UX through better APIs and indexing.\n\nBuilding on the Accelerator foundation, TON's [2026 roadmap](https://ton.org/en/roadmap) shifts focus from the infrastructure-level refactoring of the Accelerator era toward developer accessibility and its top priority, sub-second finality. At the protocol layer, Catchain 2.0 targets sub-second block finalization, while the Rust Node brings institutional-grade operability and resilience to the validator set.\n\nOn the developer side, the roadmap prioritizes four workstreams: (i) smart contract tooling through [Tolk 1.3](https://docs.ton.org/v3/documentation/smart-contracts/tolk/overview) and toolchain, the successor language and SDK stack to FunC, offering TypeScript/Rust-inspired syntax and up to 40% lower gas costs; (ii) [AppKit](https://ton.org/dev/appkit), a unified application layer via development kits for apps, wallets, and payments ([TON Pay](https://ton.org/en/ton-pay-a-new-payments-layer)), compressing the path from idea to shipped Telegram Mini App; (iii) vibe-coding workflows that let developers describe an app to an AI agent and receive a working prototype, ready to share on Telegram; and (iv) AgentKit, an MCP-based toolkit giving autonomous AI agents structured access to wallets, transfers, and DeFi modules on TON.\n\n## **Closing Summary**\n\nTON is positioning itself as one of the few L1s explicitly engineered for consumer-scale adoption, not just DeFi-native throughput benchmarks. Its core bet is that mainstream crypto use breaks on two constraints, distribution and latency/cost, and that solving both requires more than a fast chain. TON’s dynamically sharded, asynchronous architecture is designed to maintain stable performance under load, while its vertically integrated protocol services (DNS, Storage, Payments, Proxy, Sites) reduce reliance on third-party infrastructure. Most importantly, TON’s deep integration with Telegram collapses the traditional crypto UX stack (wallet → browser → dApp) into a single messaging-native environment where onchain actions can feel like normal in-app behavior.\n\nIn 2025, TON shifted to a technically differentiated L1 focused on consumer distribution, with Telegram serving as its primary onboarding and application surface. The ecosystem has increasingly matured into a Telegram-native financial stack while network activity reflects a chain transitioning from hype-driven spikes to a steadier transactional baseline.\n\nLooking forward, TON's 2026 roadmap shifts from infrastructure refactoring to performance and developer accessibility. Catchain 2.0 targets sub-second finality, the Rust Node reimplements the validator stack, and a unified developer layer (Tolk, AppKit, TON Pay) compresses the path from idea to shipped Telegram Mini App. In parallel, Telegram is becoming a native interface for AI agents, with Cocoon providing decentralized compute and AgentKit connecting autonomous agents to onchain actions. If the vibe-coding loop works at scale, where builders generate working prototypes from a natural-language prompt, share them inside Telegram for instant feedback, and iterate daily rather than quarterly, TON becomes not just a settlement layer for digital finance inside Telegram, but the fastest path from idea to testable product in crypto.",
"hook": "TON has quietly assembled the ingredients most Layer-1s lack: 1B+ captive users via Telegram, $550M+ in institutional capital, and a financial stack (stablecoins, tokenized equities, yield products) embedded directly in the messenger. This IoC covers the architecture, tokenomics, ecosystem, and 2026 roadmap driving TON's bet that consumer crypto adoption starts inside the chat window.",
"publishDate": "2026-04-07T16:00:00Z",
"readingTimeInMinutes": 21.80333333333333,
"slug": "understanding-ton-a-comprehensive-overview",
"subscriptionTier": "free",
"summary": "TON is the only Layer-1 with native distribution inside a 950M MAU messaging app. This Initiation of Coverage breaks down the network's architecture, validator economics, DeFi ecosystem, tokenomics, and 2026 roadmap. At ~$6.9B FDV, the question is whether Telegram's user base converts into sustained onchain activity.",
"tags": [
{
"id": "8ef53cf6-d66a-48e3-bf0d-03da8cfdca14",
"name": "Consumer"
},
{
"id": "9d1f93f6-d15c-40ea-b384-58c2a9419fd8",
"name": "Layer-1"
}
],
"title": "Understanding TON: A Comprehensive Overview",
"previewImage": "https://cdn.sanity.io/images/2bt0j8lu/production/cf321d93bc0b99464d29d02c79b4eda23941bfeb-6400x3600.png"
},
{
"id": "ce62bc98-3825-479a-b419-18b33d4ad5f0",
"createdAt": "2026-04-02T20:16:42Z",
"updatedAt": "2026-04-02T20:44:02Z",
"assetIds": [
"9cc76ef5-f68f-45b3-930d-98af8f6513fc"
],
"assets": [
{
"id": "9cc76ef5-f68f-45b3-930d-98af8f6513fc",
"name": "Sai",
"symbol": "Sai",
"slug": "sai-dot-fun"
}
],
"authors": [
{
"id": "cbe1f170-343a-461c-abd3-9ceee66022b8",
"name": "Eric Manoukian",
"image": "https://cdn.sanity.io/images/2bt0j8lu/production/d311e4315bfaa623b347b8ad6d1ce1e4f48023ff-400x400.jpg?w=100",
"linkedinUrl": ""
}
],
"content": "## Key Insights\n\n* Sai is a perpetual decentralized exchange (perp DEX) built on Nibiru that launched publicly on Feb. 18, 2026. It offers up to 100x leverage with oracle-settled pricing and pools liquidity through Sai Liquidity Positions (SLPs) that back markets and absorb trader P&L.\n* The launch coincided with Let's Go Saicho, a $25,500 trading competition split into two phases. Phase 1 (Feb. 18 to March 4) rewarded the top 25 traders by percentage ROI with up to $20,000, and Phase 2 (March 5 to March 19) distributes $5,500 across volume-based tracks.\n* As of March 16, TVL reached $47,211, cumulative perp volume hit $6 million, and the platform collected $21,560 in fees. Open interest peaked at $549,628 on March 3, the same day volume reached its single-day high of $1.27 million.\n* Sai saw 134 unique traders, with 52.9% executing six or more trades. Returning traders accounted for 58.2% of Phase 2 daily activity. The liquidation rate dropped from 12.9% in Phase 1 to 6.7% in Phase 2 despite average leverage increasing from 33.1x to 44.2x.\n* The roadmap extends beyond perps: a white-label Perps-as-a-Service product, Sai Savings with ~5% yield on idle stables, automated strategy vaults, real-world asset markets, and a mobile application positioning Sai as trading infrastructure rather than an isolated venue.\n\n## Primer\n\n[Sai](https://sai.fun/) is a perpetual decentralized exchange (perp DEX) that aims to simplify, predict, and make leveraged trading more accessible. Users [connect](https://docs.sai.fun/guides/wallet-setup) to the application with standard EVM wallets, deposit collateral such as USDC or stNIBI, and trade perp markets through an interface that resembles a centralized futures exchange. They can open long or short positions, set up to 100x leverage on certain assets, and use familiar order types, including market, limit, stop, and conditional (stop-loss/take-profit) orders.\n\nSai was created by members of the Nibiru team with experience in distributed systems, infrastructure, and financial protocols. The creators are led by Nibiru founder and CEO [Unique Divine](https://www.linkedin.com/in/unique-divine/), who has a background in applied mathematics and machine learning, as well as prior experience at [IBM](https://www.ibm.com/us-en) and [Sommelier Protocol](https://somm.finance/).\n\nSai is built on [Nibiru](https://nibiru.fi/) Chain, a high-throughput Layer-1 with a unified EVM and [Wasm execution](https://nibiru.fi/docs/concepts/wasm/) environment. Nibiru combines an EVM-equivalent runtime ([Nibiru EVM](https://nibiru.fi/docs/evm/)) and a Wasm environment in a single state machine, allowing Solidity and Rust contracts to coexist, share accounts and gas, and call each other through built-in system contracts that bridge the two VMs. Its consensus layer, [Nibiru BFT](https://nibiru.fi/docs/concepts/arch/nibiru-bft/), is an evolution of [CometBFT](https://docs.cometbft.com/v0.38/), delivering fast finality and sub-two-second settlements. Alongside core modules for staking, governance, IBC, and a native oracle, this architecture provides Sai with the low-latency execution, deterministic settlement, and oracle support necessary to offer a CEX-like onchain derivatives platform. Sai’s defining characteristic is its approach to pricing and risk: execution is anchored to decentralized oracles and backed by pooled vaults called Sai Liquidity Positions (SLPs), emphasizing consistent behavior across different market regimes.\n\n## Let’s Go Saicho: The Trading Competition\n\nSai's [public launch](https://x.com/SaiDotFun/status/2024080216474497407?s=20) on Feb. 18, 2026, coincided with the start of [Let's Go Saicho](https://docs.sai.fun/resources/blogs/lets-go-saicho), a one-month trading competition designed to bootstrap both trading activity and platform liquidity. The competition runs through March 19 and distributes a $25,500 prize pool across two phases, each targeting a different type of trader behavior.\n\n### Phase 1: PnL Competition\n\n[Phase 1](https://docs.sai.fun/resources/blogs/lets-go-saicho#phase-1-pnl-competition-feb-18-march-4) ran from Feb. 18, 2026, to March 4, 2026, and allocated $20,000 to the top 25 traders ranked by percentage ROI rather than absolute profit. A trader who earned a 50% return on a $500 account could outrank a trader who earned 5% on a $50,000 account. Only closed positions counted toward PnL, and unrealized gains were excluded.\n\nEligibility thresholds scaled with rank. Traders competing for the top three positions needed at least $1 million in cumulative volume and $250 in profit. Ranks 4 through 10 required $50,000 in volume and $50 in profit, while ranks 11 through 25 required $50,000 in volume with no minimum profit. The prize distribution was structured as follows:\n\n* Rank 1: $6,250\n* Rank 2: $3,125\n* Rank 3: $1,250\n* Rank 4-10: $625 each\n* Ranks 11-25: $250 each\n\n\n\nNo [traders](https://app.sai.fun/leaderboard/) met the $1 million volume threshold required for the top three positions, so none of the top three prizes were distributed. The remaining prizes were awarded to qualifying traders in ranks 4 through 25.\n\n### Phase 2: Volume Competition\n\n[Phase 2](https://docs.sai.fun/resources/blogs/lets-go-saicho#phase-2-volume-based-march-5-march-19-live-now) ran from March 5, 2026, to March 19, 2026, and shifted the incentive from profitability to volume, distributing $5,500 across three tracks. The largest pool, $4,000, is shared among all traders who cross $50,000 in volume, with each trader's share proportional to their total volume. All activity from Phase 1 counts toward Phase 2 thresholds in this segment. A $1,000 pool is split evenly among the first 50 traders to reach $10,000 in Phase 2 volume, creating an early-mover incentive. A single $500 prize goes to the highest-volume trader in Phase 2.\n\nUnlike Phase 1, Phase 2 did not require profitability to qualify. The competition enforces rules against sybil attacks, wash trading, fake volume, and malicious bots. A minimum position holding time of 10 to 20 minutes applies during Phase 2 to discourage instantaneous round-trip trades designed to inflate volume. Winners are expected to be announced shortly.\n\n## Early Traction and Key Metrics\n\n### Volume\n\n\n\nCumulative [perpetual volume](https://defillama.com/protocol/sai?tvl=false) from Feb. 18, 2026, through March 16, 2026, reached $6 million. Phase 1 accounted for $3.1 million of that total, with the final three days of Phase 1 generating $2.4 million as traders pushed to lock in PnL rankings. Phase 2 volume through March 16, 2026, totaled $2.9 million across 12 days, a higher daily average than Phase 1's first 10 days but below the Phase 1 closing sprint. Volume spiked again on March 11, 2026, at $573,420 before tapering to $36,810 on March 16, 2026, the lowest daily figure since launch, as Phase 2 approached its final days.\n\n### TVL and Open Interest\n\n\n\nAs of March 16, 2026, Sai's [total value locked](https://defillama.com/protocol/tvl/sai) (TVL) was at $47,210, up 14x from $3,380 on launch day. Growth came in two distinct waves. TVL held relatively steady between $3,200 and $4,400 during Sai's first week, then increased to $15,140 on Feb. 25, 2026, and climbed to $26,650 by Feb. 28, 2026, as early competition activity drew deposits. A second leg carried TVL from $29,120 on March 5, the start of Phase 2, to $47,210 by March 16, a 62% increase over 12 days, driven in part by new SLP deposits entering the platform after Phase 1 concluded.\n\n[Open interest](https://defillama.com/protocol/sai?openInterest=true) peaked at $549,630 on March 3, the same day perp volume hit its single-day high of $1.27 million. By March 16, open interest had settled to $27,620 as the competitive intensity of Phase 1's final days gave way to Phase 2's steadier volume-building pace.\n\n### Fees\n\n\n\nSai collected $21,560 in cumulative [fees](https://defillama.com/protocol/sai?tvl=false&perpVolume=false&fees=true) from launch through March 16. Phase 1 generated $19,600, and Phase 2 contributed $1,960 through its first 12 days. Daily fees peaked at $3,310 on Feb. 27, 2026, during a stretch from Feb. 24, 2026, to March 1, 2026, where fees exceeded $1,000 every day. After the Phase 2 transition, daily fees dropped sharply and remained subdued, falling below $100 on four of the first five days as trading shifted from high-conviction PnL plays to lower-cost volume accumulation.\n\nThe effective fee rate, fees as a percentage of volume, averaged 0.36% across the full period but diverged between phases at 0.64% in Phase 1 versus 0.07% in Phase 2. This gap likely reflects differences in sizing and trade mechanics. Phase 1 traders opened larger positions and held them through wider price moves to accumulate PnL, generating more fee-bearing events per dollar of volume. Phase 2 traders used smaller positions with higher leverage, producing high notional volume relative to the fees collected.\n\n### User Activity and Trade Frequency\n\n\n\nAs of March 16, 2026, Sai had 134 [unique traders](https://dune.com/queries/6825342/10709246). The platform had an existing base of 39 traders from its private launch period before Feb. 18, 2026, and the Let’s Go Saicho competition brought in 95 additional participants, 62 during Phase 1 and 33 during Phase 2.\n\n\n\nThe trader frequency [distribution](https://dune.com/queries/6825355/10709839) reveals a relatively engaged base for a new platform. Only 3.4% of traders executed 1 trade. The largest cohort, 43.6%, executed 2 to 5 trades, while 35.9% places 6 to 20 trades. The most active decile, 10 traders at 50+ trades each, averaged 11.8 active trading days, indicating a core group that traded consistently across both phases.\n\n### User Retention\n\n\n\n[Retention](https://dune.com/queries/6825351/10709830) patterns during Phase 2 provide a window into user stickiness. Returning traders [accounted](https://dune.com/queries/6825349/10709258) for 58.2% of daily active trader appearances from March 5, 2026, through March 16, 2026. After the March 5 spike of 60 active traders, including 37 new arrivals, daily activity settled at an average of 8.7 traders. The pattern suggests that Sai retained a meaningful share of Phase 1 participants in Phase 2 but struggled to attract new entrants after the Phase 2 launch-day momentum passed. \n\n### Risk Behavior: Liquidations and Leverage\n\nThe shift in competition incentives produced a clear change in risk behavior. The [liquidation rate](https://dune.com/queries/6825407/10709162) dropped from 12.9% of trades in Phase 1 to 6.7% in Phase 2, a 48% relative decline. This occurred despite [average leverage](https://dune.com/queries/6825407/10709162) increasing from 33.1x to 44.2x. Phase 1’s ROI-based rewards incentivized directional bets with higher risk tolerance, meaning getting liquidated was an acceptable cost to pursue outsized returns. Phase 2’s volume-based rewards penalized liquidations indirectly, since blown-up capital can’t generate more volume.\n\n## Community: The SaiClone Ambassador Program\n\nAlongside the trading competition, Sai launched the [SaiClone Ambassador Program](https://docs.sai.fun/resources/blogs/saiclone-ambassador), a three-tier progression system that operates entirely through Sai's [Discord](https://discord.com/invite/saidotfun) server. The program uses the [Mee6](https://mee6.xyz/en) bot to track contributions and assign XP, rewarding community engagement, content creation, and platform advocacy across the following tiers:\n\n* [Saicho](https://docs.sai.fun/resources/blogs/saiclone-ambassador#id-1.-saicho) (levels 0-5, up to 1,624 XP): The entry tier, granted automatically upon engaging with the Discord community.\n* [Saiborg](https://docs.sai.fun/resources/blogs/saiclone-ambassador#id-2.-saiborg) (Levels 6-15, 1,625-13,799 XP): Trusted, highly engaged members who receive special community recognition and increased influence on community decisions.\n* [Sage](https://docs.sai.fun/resources/blogs/saiclone-ambassador#id-3.-sage) (Level 16+, 13,800+ XP): Reserved for top contributors. Unlike other tiers, Sage requires a formal application reviewed by the Sai team. Benefits include direct team access, exclusive merchandise, bi-monthly raffles, collaboration opportunities, and early access to platform updates.\n\n[XP](https://docs.sai.fun/resources/blogs/saiclone-ambassador#how-to-earn-xp) accrues through trading activity, technical analysis and signals, community events, video and educational content, X engagement, bug reports, and Discord messages. The program ties community growth directly to platform usage, creating a feedback loop between trading activity and ambassador progression.\n\n## Looking Ahead\n\nSai's roadmap positions the platform as a trading infrastructure rather than a standalone venue. The most differentiated planned product is a white-label Perps-as-a-Service offering, with the first [iteration](https://x.com/SaiDotFun/status/2032458407899209822) from [Coded Estate](https://app.codedestate.com/perps/trades), an RWA platform that now offers perpetual trading powered by Sai. If executed, this would extend Sai's liquidity and infrastructure beyond its own front end.\n\nOn the product side, planned additions include [Sai Savings](https://docs.sai.fun/resources/blogs/intro-to-sai#new-products-and-yield:~:text=products%20and%20yield-,Sai%20Savings,-%3A%20A%20way%20for), which targets ~5% yield on idle stable balances, [automated strategy vaults](https://docs.sai.fun/resources/blogs/intro-to-sai#new-products-and-yield:~:text=control%20and%20flexibility-,Automated%20strategies,-%3A%20Launch%20vaults), expanded [market listings](https://docs.sai.fun/resources/blogs/intro-to-sai#new-products-and-yield:~:text=Real%20world%20asset%20markets) that include real-world asset-style instruments, and DeFi integrations connecting SLP vaults to swaps and routing protocols. The [account](https://docs.sai.fun/resources/blogs/intro-to-sai#new-products-and-yield:~:text=with%20one%20click-,Accounts%3A,-A%20true%20CEX) experience is also evolving. Gasless trades are already live, and the team plans to add multichain and fiat funding routes, [cross-chain USDC deposits](https://docs.sai.fun/resources/blogs/intro-to-sai#new-products-and-yield:~:text=Cross%20chain%20and%20fiat%20funding) via improved on-ramps, and a [mobile application](https://docs.sai.fun/resources/blogs/intro-to-sai#new-products-and-yield:~:text=or%20bank%20account-,Mobile%20app%3A,-Release%20a%20mobile).\n\nFor developers, Sai plans to ship a [data platform](https://docs.sai.fun/resources/blogs/intro-to-sai#new-products-and-yield:~:text=the%20Sai%20platform-,Data%20platform,-%3A%20Backtesting%2C%20historical%20data) with historical data access, backtesting tools, and support for custom order types.\n\n## Closing Summary\n\nSai's Let's Go Saicho trading competition reached 134 traders, generated $6 million in volume, grew TVL 14x in under a month, and collected $21,560 in fees. The two-phase incentive structure produced distinct behavioral shifts, with liquidation rates halving and position sizes shrinking as traders adapted from ROI-focused to volume-focused trading. A 58.2% retention rate in Phase 2 suggests the platform established a sustained baseline of recurring users, though new trader acquisition dropped sharply after the Phase 2 launch spike.\n\nThe next phase for Sai begins after the competition ends. Whether the platform can retain its active traders without incentives, grow SLP liquidity organically, and begin delivering on a roadmap spanning Perps-as-a-Service, Sai Savings, real-world asset markets, and a mobile application will determine whether Sai converts early traction into lasting positioning within the perp DEX landscape.",
"hook": "Sai launched on Nibiru with a two-phase trading competition that attracted 134 traders and generated $6 million in perpetual volume in under a month.",
"publishDate": "2026-04-06T13:50:00Z",
"readingTimeInMinutes": 10.556666666666667,
"slug": "sai-bootstrapping-a-perp-dex-on-nibiru",
"subscriptionTier": "free",
"summary": "Sai is a perpetual DEX on Nibiru offering up to 100x leverage with oracle-settled pricing and pooled SLP-backed liquidity. Its public launch coincided with Let's Go Saicho, a $25,500 competition split between a PnL phase and a volume phase, each targeting different trader behavior. TVL grew 14x to $47,210, cumulative fees hit $21,560, and returning traders accounted for 58.2% of Phase 2 daily activity. The effective fee rate diverged sharply between phases, from 0.64% in Phase 1 to 0.07% in Phase 2, reflecting a shift from high-conviction directional bets to smaller, higher-leverage volume plays. Sai's roadmap extends into Perps-as-a-Service, yield products, RWA markets, and a mobile app, positioning it as trading infrastructure rather than an isolated venue.",
"tags": [
{
"id": "8f138ee0-fb4f-42b8-919c-9feffd558d88",
"name": "Perp DEX"
},
{
"id": "3fdf6b6b-cfac-426d-b20f-3e4794dec982",
"name": "Pulse Reports"
}
],
"title": "Sai: Bootstrapping a Perp DEX on Nibiru",
"previewImage": "https://cdn.sanity.io/images/2bt0j8lu/production/a4aec153be082d301a6690dc37248c9c4ed649d5-6068x3560.png"
},
{
"id": "283895eb-251b-416c-99b1-ed5c3cb67f9c",
"createdAt": "2026-03-29T21:09:41Z",
"updatedAt": "2026-03-31T16:11:30Z",
"assetIds": [
"3748291f-99cb-4002-9bf2-1a23edc0af77"
],
"assets": [
{
"id": "3748291f-99cb-4002-9bf2-1a23edc0af77",
"name": "Nexus Labs",
"symbol": "NEX",
"slug": "nexus-labs"
}
],
"authors": [
{
"id": "25fb4f17-1b40-4477-8e6b-be982a87b349",
"name": "Alice Hou",
"image": "https://cdn.sanity.io/images/2bt0j8lu/production/57dad35450be1a47f2ff6628e6f5fca213499ab4-400x400.jpg?w=100",
"linkedinUrl": ""
},
{
"id": "ac138b81-4baf-4a63-8dbc-e8592c93ba17",
"name": "Matt Kreiser",
"image": "https://cdn.sanity.io/images/2bt0j8lu/production/a09ff5484ce95234175e5e49131ab71bd1b1726d-1024x1024.png?w=100",
"linkedinUrl": ""
}
],
"content": "## Key Insights\n\n* Nexus embeds **high-performance financial engines directly into the protocol through its co-processor model**, moving exchange, margin, and liquidation logic from contract-level simulation into native execution.\n* The **dual-execution architecture allows performance-critical financial workloads and programmable smart contracts to operate in parallel**, avoiding the typical tradeoff between latency and composability.\n* **The Nexus zkVM anchors execution to cryptographic proofs** rather than full validator re-execution, positioning **proof verification as the primary mechanism for scalable correctness.**\n* **With mainnet and exchange deployment in 2026**, Nexus shifts from infrastructure buildout to market validation, where **liquidity formation, and proof-generation efficiency will determine long-term viability. Central to achieving market validation are USDX, the native U.S. dollar stablecoin of the Nexus ecosystem, and the Nexus Exchange, a non-custodial, central limit order book (CLOB) embedded directly into the Nexus Layer 1 (L1).**\n* **The Nexus Exchange is designed to deliver CEX-parity performance in a more self-custodial, verifiable environment that can host and leverage high-frequency trading strategies, AI agents, commerce, and other economic activity.**\n\n## Introduction\n\nFinancial infrastructure depends on complex computational systems that remain largely unverifiable to external participants. Risk models, margin calculations, settlement logic, and internal reconciliation processes operate behind institutional boundaries, requiring users to rely on reporting and oversight rather than direct proof. While public blockchains introduced deterministic execution and transparent state transitions, most existing architectures are not designed to support the performance and computational demands of modern financial markets.\n\nGeneral-purpose chains prioritize composability and shared liquidity but face constraints around latency and throughput. [Application-specific chains](https://messari.io/copilot/share/application-specific-chains-d7e77c8d-2068-4560-84a1-c23e800490dc) achieve higher performance by narrowing scope, yet fragment liquidity and isolate execution environments. In both cases, critical financial logic often remains either offchain or insufficiently optimized for high-frequency, computation-heavy workloads.\n\n[Nexus](https://messari.io/project/nexus-labs) introduces a layered architecture that separates execution, verification, and consensus into independently optimized systems. Its [dual execution model](https://docs.nexus.xyz/architecture/dual-block-execution) combines an [EVM-compatible environment](https://messari.io/copilot/share/evm-compatibility-df970104-0577-49f5-a15e-647f9e746f32) with a specialized financial [co-processor](https://messari.io/copilot/share/co-processor-definition-6dd4fd0e-6438-4882-8403-5fa2234b1ae2), while a native [zkVM](https://messari.io/copilot/share/zkvm-explained-81e8af21-442d-44b7-9201-1a191baf51ef) generates proofs of correct execution that are committed to the base layer. This structure is designed to support performance-sensitive financial applications without relying on external verification frameworks. For example, the [upcoming](https://blog.nexus.xyz/introducing-the-nexus-dex-alpha/) [Nexus ](https://app.nexus.xyz/trade)Exchange is designed to deliver CEX-parity performance in a more decentralized environment that can host and leverage high-frequency trading strategies, AI agents, commerce, and other economic activity.\n\nThis Initiation of Coverage (IOC) report focuses on a technical examination of Nexus’s architecture. Readers seeking a broader discussion of the long-term vision behind verifiable finance are encouraged to refer to the [Nexus Pulse Report](https://messari.io/report/nexus-a-framework-for-verifiable-finance).\n\n[Website](https://nexus.xyz/) / [X (Twitter)](https://x.com/NexusLabs) / [Discord](https://discord.com/invite/AmRKShJfq6)\n\n## Background\n\n[Nexus](https://messari.io/project/nexus-labs/fundraising) was founded in 2022 by [Daniel Marin](https://www.linkedin.com/in/danielmarinq/), a computer science graduate from Stanford University, with the objective of building a universal verifiable machine capable of proving arbitrary computation. The project initially focused on advancing zero-knowledge (zk) proving infrastructure before refining its scope toward financial applications that demand both high performance and computational integrity.\n\nIn June 2024, Nexus [raised](https://blog.nexus.xyz/series-a/) a $25 million Series A round co-led by [Lightspeed](https://messari.io/organization/lightspeed-venture-partners) and [Pantera](https://messari.io/organization/pantera-capital), bringing total capital raised to over $27 million. The network [launched](https://blog.nexus.xyz/the-new-nexus-testnet-is-live/) its first public testnet in December 2024 and has since iterated toward a production-ready architecture.\n\n## Technology\n\n\n\nNexus is structured as a [three-layer architecture](https://docs.nexus.xyz/) composed of an Execution Layer, a Verification Layer, and a Consensus Layer. Each layer operates as a distinct system, responsible for running application logic, generating and validating execution proofs, and finalizing state transitions.\n\n### Execution Layer: Nexus EVM and NexusCore\n\n#### Dual-Block Execution\n\nNexus implements a [dual-block execution model](https://docs.nexus.xyz/architecture/dual-block-execution) designed to separate high-frequency financial processing from general-purpose smart contract coordination. Instead of batching all activity into a single block cadence, the network operates two synchronized block streams with distinct performance characteristics.\n\n[NexusCore](https://docs.nexus.xyz/architecture/nexuscore) targets block times of five milliseconds, while NexusEVM has a block time of 1-2 seconds. These blocks are optimized for latency-sensitive workloads such as order matching, position updates, and risk recalculations. [NexusEVM](https://docs.nexus.xyz/architecture/nexusevm) operates on a slower block cadence, aggregating state changes from several NexusCore blocks before finalization. This periodic synchronization layer preserves compatibility with Ethereum-style smart contracts while allowing composability between programmable logic and high-speed financial activity.\n\nThis structure has three key effects. First, high-speed trading activity runs independently from complex smart contract logic, so time-sensitive operations are not slowed down by heavier computation. Second, performance can scale more efficiently with hardware, as financial workloads do not need to wait for full EVM block processing. Third, both execution environments remain economically unified, with fees and incentives settled at the base layer rather than split across separate systems.\n\nNexusCore\n\nAt the center of NexusCore is the co-processor model. A co-processor is a native execution module embedded into the blockchain itself. Instead of interpreting [smart contract bytecode](https://blog.chain.link/what-are-abi-and-bytecode-in-solidity/), it runs pre-compiled logic with direct access to protocol resources. This design reduces execution overhead and allows financial operations to be processed with greater consistency and speed.\n\nEach co-processor functions as an independent state machine. It maintains its own isolated data structures, executes specialized algorithms tailored to its purpose, and exposes controlled interfaces for interaction. This isolation enables parallel execution across modules while maintaining deterministic state transitions under shared consensus validation.\n\n\n\nThe architecture can be understood in three components:\n\n* **State Layer:** Maintains dedicated data structures and deterministic updates for each co-processor.\n* **Machine Layer:** Executes specialized financial logic, such as matching, margining, or settlement.\n* **I/O Layer:** Enables co-processors to be accessed directly by offchain systems or by smart contracts onchain. \n\nNexusCore’s dual interface is a key advantage. It allows professional trading systems to connect directly for speed, while smart contracts can interact with the same engine onchain. In practice, both high-frequency trading and DeFi applications can run on the same network without sacrificing performance or composability.\n\nOver time, NexusCore is intended to host a broader catalog of “L1-native engines,” including lending markets, vault strategies, oracle and information feeds, RWA and stablecoin infrastructure, gas and fee modules, and bridging primitives, all of which compose atomically with NexusEVM smart contracts.\n\n#### NexusEVM\n\nAt the protocol level, NexusEVM adheres to the standard [Ethereum Virtual Machine ](https://ethereum.org/developers/docs/evm/)specification. It supports the same contract bytecode, gas semantics, RPC interfaces, and developer tooling used across Ethereum.\n\n\n\nWithin Nexus, NexusEVM runs in parallel with NexusCore. Smart contracts deployed on NexusEVM can invoke Core-level co-processors through EVM precompiles or structured cross-domain calls. These interactions are [atomic](https://www.investopedia.com/terms/a/atomic-swaps.asp), meaning that if any part of the transaction fails, the entire operation reverts. Ordered processing ensures deterministic state transitions across validators.\n\nThis integration allows developers to combine programmable contract logic with high-performance financial engines. Applications can manage governance, token logic, incentives, or strategy layers in NexusEVM, while delegating performance-critical execution to NexusCore. In practice, NexusEVM provides an expressive, composable layer of the system, enabling developers to extend and build on top of Core-level financial primitives without leaving the base chain.\n\n### Verification Layer: Nexus zkVM\n\nThe Verification Layer is powered by the Nexus zkVM, a zk virtual machine that generates proofs confirming that computation was executed exactly as specified. Instead of every validator replaying complex logic, the network verifies a succinct proof derived from that execution.\n\nThe zkVM is composed of four primary technical layers:\n\n* RISC-V Machine Architecture: A custom-built virtual machine implementing a modified [RISC-V instruction set](https://messari.io/copilot/share/risc-v-b8a3d4a4-438c-487a-a4ad-53fde178e4f0). It is designed specifically for prover efficiency, including structured memory handling and a “prove only what is accessed” model that reduces unnecessary proof overhead.\n* Algebraic Constraint System (AIR): The execution of the machine is translated into a mathematical representation known as an [Algebraic Intermediate Representation](https://docs.nexus.xyz/zkvm/overview/architecture#core-components). This formalizes every instruction, memory read, and state transition into constraints that must be satisfied for a proof to be valid.\n* STARK-Based Prover (S-two Integration): Execution traces are converted into cryptographic proofs using a [STARK prover](https://github.com/starkware-libs/stwo) optimized for performance. STARKs allow proofs to remain succinct and publicly verifiable without trusted setup requirements.\n* Runtime & SDK Layer: A [Rust](https://rust-lang.org/)-based runtime that allows developers to define public inputs, private inputs, outputs, and logging in a structured way, while abstracting the underlying proving complexity.\n\nNexus zkVM’s benefit is architectural scalability: computation can scale independently from consensus because validators verify proofs rather than replay entire workloads. This reduces replication cost while maintaining deterministic correctness.\n\nA potential risk, on the other hand, is proving overhead. Generating STARK proofs is computationally intensive and requires specialized hardware or distributed prover infrastructure. While verification is lightweight, the economic viability of large-scale proving depends on continued optimization and network-level prover coordination.\n\n### Consensus Layer: Nexus BFT\n\nThe Consensus Layer is governed by [NexusBFT](https://docs.nexus.xyz/), the protocol responsible for finalizing blocks, validating execution commitments, and managing the lifecycle of co-processors.\n\nEach block finalized by NexusBFT includes three core elements:\n\n* A [Merkle commitment](https://messari.io/copilot/share/merkel-commitment-d677ea83-4c90-41b6-b851-bd75041acea5) to the execution state, anchoring the verified results of both NexusCore and NexusEVM.\n* Validator signatures and metadata, establishing agreement across the network.\n* Optional registry updates, which modify the active set of co-processors through the [CPRegistry](https://docs.nexus.xyz/network/overview/system-overview#consensus-layer).\n\nBeyond standard block finalization, NexusBFT introduces protocol-level extensibility. Rather than requiring [hard forks](https://messari.io/copilot/share/what-are-hard-forks-5a6fa661-fb57-4817-9208-aac62b515239) to introduce or modify financial engines, co-processor registration and lifecycle management are handled directly within the consensus layer. This allows the network to activate, upgrade, or deprecate specialized modules without disrupting execution environments.\n\n## DeFi on Nexus\n\n### Strategy and USDX\n\nThe Nexus team is building a native stablecoin and perpetuals-focused Exchange directly into the Layer 1, given that these are [three of the most](https://blog.nexus.xyz/usdx-the-money-layer-for-verifiable-finance/#:~:text=Ethereum%2C%20and%20beyond.-,The%20USDX%20thesis,-At%20the%20core) extensible, proven and synergistic businesses in crypto. Taken together, the L1 and exchange are designed to deliver CEX-parity performance in a more decentralized environment that can host and leverage high-frequency trading strategies, AI agents, commerce, and other economic activity.\n\nUSDX is the [native U.S. dollar stablecoin](https://blog.nexus.xyz/usdx-the-money-layer-for-verifiable-finance/) of the Nexus ecosystem, 1:1-backed by U.S. Treasury bills and cash equivalents. USDX will be the default margin and settlement asset for [Nexus ](https://app.nexus.xyz/trade)Exchange, a non-custodial [central limit order book](https://messari.io/copilot/share/central-limit-order-book-0364d6bc-f9e7-4059-99e7-28066b7cc1f5) (CLOB) embedded directly into the Layer 1 (L1). The more the Exchange is used, the greater the demand for USDX. Moreover, the more USDX in circulation, the deeper and cheaper liquidity becomes for the exchange. With meaningful adoption, this reinforcing flywheel can increase the value of the L1.\n\nUSDX will first launch on Ethereum, where distribution is deepest, followed by Nexus mainnet, and crosschain interoperability. Issuance will take place via the M0 Protocol via the JMI extension, which will enable permissionless 1:1 swaps from major stablecoins into USDX. As tokenized assets and 24/7 synthetic markets expand onchain, USDX is designed to serve as the neutral denominator for risk-managed portfolios.\n\nThe key value proposition of USDX relative to other stablecoins is [yield streaming](https://blog.nexus.xyz/yield-is-the-2026-defi-battleground/#:~:text=DeFi%20in%202026.-,Yield%20streaming%3A%20from%20extraction%20to%20alignment,-Yield%20streaming%20inverts), which inverts the [extractive model](https://blog.nexus.xyz/yield-is-the-2026-defi-battleground/#:~:text=The%20extractive%20model%3A) of the largest stablecoins USDT and USDC, whose issuers capture all reserve yield as compensation for infrastructure and compliance. Nexus streams USDX yield directly to users and builders on the L1 via the fully transparent Global Yield Distribution System ([GYDS](https://blog.nexus.xyz/usdx-the-money-layer-for-verifiable-finance/#:~:text=Nexus-,Yield%20streaming%20and%20the%20Global%20Yield%20Distribution%20System,-One%20of%20the)). Yield from U.S. Treasury bills and cash equivalents is distributed as USDX each week according to the time-weighted USDX balances across registered application sources, which are contracts and modules that builders have opted into. The yield is split between the protocol and builders, with builder allocations proportional to attributed USDX TVL, and yield flowing through to end users based on their balances, where supported. On top of this, Nexus can, at its discretion, add protocol-native incentives to increase yield beyond short-duration U.S. Treasury yields, including portions of exchange revenue, subject to governance and risk budgets, directed at strategic segments like builders bringing new markets or liquidity. The Nexus team plans to publish a quarter-by-quarter policy for USDX, disclosing a target onchain yield that reflects the baseline Treasury rate plus any protocol-native additions.\n\nThis aligns incentives to a much higher degree than Tether’s USDT and Circle’s USDC, which simply offer access to the stablecoin as the value proposition. It’s also a process entirely unblocked compared to yield-bearing stablecoins registered as securities (which severely limits the holder base to qualified participants) in order to pay interest directly to holders. \n\nThe yield streaming model of USDX creates a predictable, programmatic incentive loop. Builders are rewarded for attracting real USDX usage, while users are rewarded for holding and deploying USDX on Nexus. During the early phase, NEX token incentives may complement USDX yield to accelerate integrations, but the long-term design relies on organic demand and real yield.\n\n### Nexus Exchange\n\nTo operationalize its execution architecture, Nexus [is developing](https://blog.nexus.xyz/introducing-the-nexus-dex-alpha/) the [Nexus ](https://app.nexus.xyz/trade)Exchange, a non-custodial [central limit order book](https://messari.io/copilot/share/central-limit-order-book-0364d6bc-f9e7-4059-99e7-28066b7cc1f5) (CLOB) embedded directly into the Layer 1 (L1). Unlike exchanges deployed as smart contracts, the Nexus Exchange runs inside NexusCore as a native co-processor. Order matching, margin calculations, funding logic, and liquidations are executed at the protocol level rather than simulated through contract bytecode.\n\nThe first supported product is [perpetual futures](https://docs.nexus.xyz/trading/perpetuals). These contracts allow traders to take leveraged long or short exposure to supported assets without expiration. Users post collateral, open positions with leverage, and pay or receive periodic funding based on market conditions. Because the exchange engine operates within NexusCore, all position updates and risk calculations are processed under deterministic execution and shared consensus.\n\nThis architecture shifts the exchange from being an application layered on top of the chain to becoming part of the chain’s execution fabric. Performance-sensitive operations are handled natively, while settlement and accounting remain transparent and verifiable at the base layer.\n\n### Risk Management and Liquidation\n\nRisk controls are enforced automatically through a built-in [liquidation engine](https://docs.nexus.xyz/trading/perpetuals/liquidations). A trader’s equity is continuously evaluated against predefined margin thresholds using a mark price derived from the Nexus oracle system.\n\nTwo margin levels govern positions. Initial margin determines the leverage required to open a position, while maintenance margin defines the minimum equity needed to keep it open. If equity falls below the maintenance threshold, the system triggers liquidation.\n\nRather than relying on external bots competing to liquidate positions, Nexus executes liquidations through a dedicated onchain mechanism. Positions are closed at or near the mark price, with safeguards to prevent negative balances. If losses exceed available collateral, an [insurance fund](https://docs.nexus.xyz/trading/perpetuals/liquidations#how-liquidation-works) absorbs residual shortfalls to preserve overall system solvency.\n\nBy embedding margin and liquidation logic directly into the execution layer, Nexus reduces execution uncertainty and race conditions. The tradeoff is that core risk parameters are embedded at the protocol level, making changes more consequential than in contract-based systems.\n\nTogether, the Nexus Exchange functions as both a flagship application and a structural demonstration of NexusCore’s capabilities. It tests whether protocol-level financial engines can combine high-performance execution with deterministic settlement and cryptographic accountability within a unified L1 environment. Beyond perpetual futures, Nexus plans to expand the Exchange to include [spot markets](https://docs.nexus.xyz/trading/spot) and [vault products](https://docs.nexus.xyz/trading/vaults), with further details expected in future releases.\n\n## Roadmap \n\nNexus’s [roadmap](https://blog.nexus.xyz/the-nexus-roadmap/) outlines a transition from testnet infrastructure to a fully operational financial L1, with staged activation of validators, exchange functionality, and protocol-level financial primitives.\n\n### Q1 2026: Network Activation\n\nThe first milestone in 2026 is under active development and will center on network coordination and exchange readiness. [Community Genesis](https://blog.nexus.xyz/tag/community/) will onboard validators, operators, and early participants, marking the shift from a development-driven network to a validator-secured environment.\n\n### Q2 2026: Mainnet EVM Launch\n\nIn Q2 2026, the Nexus L1 mainnet is expected to go live. This milestone will establish a production settlement layer with secure execution and finalized consensus. Bridges and onramps will become operational, allowing external capital to enter the ecosystem.\n\nThis stage will formalize validator participation and transition the EVM environment from testnet to persistent infrastructure. Applications will be able to deploy into a stable environment with deterministic finality and integrated access to NexusCore co-processors. The focus will shift from experimentation to economic durability.\n\n### Q3 2026: Exchange Mainnet\n\nFollowing L1 activation, the Nexus Exchange will launch on mainnet. This will mark the operational start of Nexus as a functioning financial network rather than solely an infrastructure layer. The exchange is designed to deliver CEX-like performance in a more decentralized environment that can host and leverage high-frequency trading strategies, AI agents, commerce, and other economic activity.\n\nLive trading will introduce continuous order flow, real margin enforcement, oracle updates under production conditions, and full interaction between execution, verification, and consensus layers. Liquidity depth, liquidation behavior, and proof performance will become measurable under sustained market activity.\n\nBeyond launch milestones, Nexus will continue to evolve its core architecture. Development efforts will focus on improving zkVM performance, expanding co-processor capabilitiesand broadening supported asset classes such as 24/7 equities, FX, commodities, and indexes.\n\nThe long-term objective will be to extend protocol-level financial infrastructure across additional markets, collateral models, and composable applications. This phase will prioritize improvements in proof efficiency, execution reliability, and validator robustness to support sustained financial activity.\n\n## Closing Summary\n\nNexus represents an architectural bet: that high-performance financial infrastructure should not be simulated through smart contracts, but embedded directly into the base layer and verified cryptographically.\n\nIts three-layer design separates execution, proof generation, and consensus, allowing each to scale independently. The dual-execution model formalizes a distinction between programmable logic and performance-critical financial engines. NexusCore handles deterministic, latency-sensitive computation, while NexusEVM preserves composability and developer accessibility. The zkVM anchors the system with verifiable computation, shifting validation from re-execution to proof verification.\n\nWith mainnet and exchange deployment in 2026, Nexus shifts from infrastructure buildout to market validation, where liquidity formation, oracle reliability, and proof-generation efficiency will determine long-term viability. Central to achieving market validation are USDX, the native U.S. dollar stablecoin of the Nexus ecosystem, and the Nexus Exchange, a non-custodial, central limit order book (CLOB) embedded directly into Layer 1 (L1). USDX inverts the extractive model of the largest stablecoins USDT and USDC, whose issuers capture all reserve yield as profit for infrastructure and compliance. Nexus streams USDX yield directly to users and builders on the L1 via the fully transparent Global Yield Distribution System (GYDS). The Nexus Exchange is designed to deliver CEX-parity performance in a more decentralized environment that can host and leverage high-frequency trading strategies, AI agents, commerce, and other economic activity.\n\nBy integrating matching, margining, liquidation, and oracle logic at the protocol level, Nexus reduces overhead and execution uncertainty inherent in contract-based exchange designs. The tradeoff is structural: financial logic becomes part of the base layer, increasing the importance of validator coordination and disciplined protocol governance.\n\nUltimately, the success of Nexus will not be measured by throughput alone, but by whether protocol-level financial primitives can operate reliably under live market conditions while maintaining deterministic settlement and cryptographic accountability.",
"hook": "Nexus embeds high-performance financial engines directly into the protocol through its co-processor model, moving exchange, margin, and liquidation logic from contract-level simulation into native execution.",
"publishDate": "2026-03-31T14:00:00Z",
"readingTimeInMinutes": 14.376666666666667,
"slug": "understanding-nexus-a-comprehensive-overview",
"subscriptionTier": "free",
"summary": "The dual-execution architecture allows performance-critical financial workloads and programmable smart contracts to operate in parallel, avoiding the typical tradeoff between latency and composability. The Nexus zkVM anchors execution to cryptographic proofs rather than full validator re-execution, positioning proof verification as the primary mechanism for scalable correctness. With mainnet and exchange deployment in 2026, Nexus shifts from infrastructure buildout to market validation, where liquidity formation, and proof-generation efficiency will determine long-term viability. Central to achieving market validation are USDX, the native U.S. dollar stablecoin of the Nexus ecosystem, and the Nexus Exchange, a non-custodial, central limit order book (CLOB) embedded directly into the Nexus Layer 1 (L1).",
"tags": [
{
"id": "2df3c948-022e-4a95-9ecf-d2ddca4c40f3",
"name": "Protocol Overview"
}
],
"title": "Understanding Nexus: A Comprehensive Overview",
"previewImage": "https://cdn.sanity.io/images/2bt0j8lu/production/6c006611ca5148c732fac20be7f0f8ecc6641368-6400x3600.png"
}
],
"metadata": {
"limit": 10,
"page": 1,
"totalRows": 2,
"totalPages": 1
}
}{
"data": "<unknown>",
"error": "<string>"
}{
"data": "<unknown>",
"error": "<string>"
}{
"data": "<unknown>",
"error": "<string>"
}{
"data": "<unknown>",
"error": "<string>"
}Research (Basic)
Get basic research reports
Returns a list of basic research reports based on filters like asset ID and tags
GET
/
research-basic
/
v1
/
basic
/
reports
Get basic research reports
curl --request GET \
--url https://api.messari.io/research-basic/v1/basic/reports \
--header 'X-Messari-API-Key: <api-key>'import requests
url = "https://api.messari.io/research-basic/v1/basic/reports"
headers = {"X-Messari-API-Key": "<api-key>"}
response = requests.get(url, headers=headers)
print(response.text)const options = {method: 'GET', headers: {'X-Messari-API-Key': '<api-key>'}};
fetch('https://api.messari.io/research-basic/v1/basic/reports', options)
.then(res => res.json())
.then(res => console.log(res))
.catch(err => console.error(err));<?php
$curl = curl_init();
curl_setopt_array($curl, [
CURLOPT_URL => "https://api.messari.io/research-basic/v1/basic/reports",
CURLOPT_RETURNTRANSFER => true,
CURLOPT_ENCODING => "",
CURLOPT_MAXREDIRS => 10,
CURLOPT_TIMEOUT => 30,
CURLOPT_HTTP_VERSION => CURL_HTTP_VERSION_1_1,
CURLOPT_CUSTOMREQUEST => "GET",
CURLOPT_HTTPHEADER => [
"X-Messari-API-Key: <api-key>"
],
]);
$response = curl_exec($curl);
$err = curl_error($curl);
curl_close($curl);
if ($err) {
echo "cURL Error #:" . $err;
} else {
echo $response;
}package main
import (
"fmt"
"net/http"
"io"
)
func main() {
url := "https://api.messari.io/research-basic/v1/basic/reports"
req, _ := http.NewRequest("GET", url, nil)
req.Header.Add("X-Messari-API-Key", "<api-key>")
res, _ := http.DefaultClient.Do(req)
defer res.Body.Close()
body, _ := io.ReadAll(res.Body)
fmt.Println(string(body))
}HttpResponse<String> response = Unirest.get("https://api.messari.io/research-basic/v1/basic/reports")
.header("X-Messari-API-Key", "<api-key>")
.asString();require 'uri'
require 'net/http'
url = URI("https://api.messari.io/research-basic/v1/basic/reports")
http = Net::HTTP.new(url.host, url.port)
http.use_ssl = true
request = Net::HTTP::Get.new(url)
request["X-Messari-API-Key"] = '<api-key>'
response = http.request(request)
puts response.read_body{
"error": null,
"data": [
{
"id": "16d6619b-9416-46cc-8fd2-13fdcd9bae2b",
"createdAt": "2026-04-06T23:37:35Z",
"updatedAt": "2026-04-07T16:00:58Z",
"assetIds": [
"2023433a-23f4-4901-822d-a537b0c71676"
],
"assets": [
{
"id": "2023433a-23f4-4901-822d-a537b0c71676",
"name": "Toncoin",
"symbol": "TON",
"slug": "the-open-network"
}
],
"authors": [
{
"id": "872be9a6-e5d6-47c4-ac6d-95bb1720adf9",
"name": "Whynonah",
"image": "https://cdn.sanity.io/images/2bt0j8lu/production/8ac8076d5648490dd40511ab3b81c602a03be811-681x793.jpg?w=100",
"linkedinUrl": ""
},
{
"id": "2b10859c-dd75-472e-afde-1d1df4d2f576",
"name": "Jonny Kreiser",
"image": "https://cdn.sanity.io/images/2bt0j8lu/production/5f0238a98921e3b37e6b3034ebe88c6561fd393c-400x400.jpg?w=100",
"linkedinUrl": ""
}
],
"content": "## **Key Insights**\n\n* **Toncoin (TON) functions as the network’s core execution asset (i.e., paying gas, supporting DeFi liquidity, securing Proof-of-Stake consensus, and acting as a settlement layer)** and as the foundation of Telegram's in-app economy.\n* **2025 activity shows TON shifting from early-year viral surges into a steadier baseline** (~100K–150K daily active users and ~1.5–2.5 million daily transactions).\n* **TON’s ecosystem** is evolving into a Telegram-native financial stack, with stablecoins, yield products, and RWAs increasingly embedded in wallets and mini-apps; **highlighted by xStocks, Ethena, Tether, Affluent, and liquidity routing via STON.fi + Omniston.**\n* **TON's 2026 roadmap pivots from infrastructure refactoring to performance and developer accessibility**. Catchain 2.0 targets sub-second finality, the Rust Node reimplements the validator stack, and a unified developer layer that includes developer kits for smart contracts, apps, wallets and payments.\n* **TON is emerging as a native infrastructure layer for AI** inside Telegram, with Cocoon providing decentralized compute, AgentKit connecting autonomous agents to onchain actions, and vibe-coding workflows enabling builders to quickly generate and share working prototypes.\n\n## **Introduction**\n\nConsumer crypto adoption typically breaks down into two bottlenecks: distribution and cost/latency. Most blockchains rely on standalone wallets and browser-based dApps, creating multi-step onboarding funnels that deter mainstream users. Even when infrastructure works, blockchain interactions often feel slow, expensive, or fragmented compared to Web2 applications; particularly in consumer use cases like payments, gaming, and social applications, where users expect instant feedback and low friction.\n\nTON’s thesis is that mass adoption requires both scalable infrastructure and native distribution. At the base layer, TON is built as a dynamically sharded Proof-of-Stake (PoS) network with asynchronous smart contracts, allowing parallel execution and horizontal scalability. Rather than relying solely on high single-chain throughput, TON’s architecture is designed to maintain stable performance under load. Complementing the Layer-1 are native protocol services, including (i) [TON DNS](https://dns.ton.org/) (human-readable “.ton” names), (ii) [TON Storage](https://docs.ton.org/foundations/services#ton-storage) (decentralized file storage), (iii) [TON Payments](https://docs.ton.org/foundations/services#ton-payments) (payment channels), (iv) [TON Proxy](https://docs.ton.org/foundations/services#ton-proxy) (censorship-resistant routing), (v) [TON Sites ](https://ton.org/en/ton-sites)(decentralized websites), and (vi) [Tolk](https://docs.ton.org/languages/tolk/overview#tolk-language) (smart contract language), and (vii) [AppKit](https://docs.ton.org/ecosystem/appkit/overview) (an all-in-one SDK for building Telegram Mini Apps with TON), which together form a vertically integrated blockchain stack with easy developer tooling.\n\nWhat differentiates TON is its distribution and tight integration with [Telegram](https://telegram.org/). [**TON Wallet**](https://wallet.tg/ton), natively embedded in the messenger, allows users to transact onchain without leaving the app. [**TON Connect**](https://docs.ton.org/ecosystem/ton-connect/overview), the wallet-connection protocol for **Telegram Mini Apps (TMAs)**, opens this to any compatible third-party wallet, though TON Wallet remains unique as Telegram's built-in wallet. Combined with in-chat app distribution, the result is onchain actions that feel like normal in-app behaviour, collapsing the traditional crypto UX stack into a single consumer environment.\n\n[Website](https://ton.org/en) / [X (Twitter)](https://x.com/ton_blockchain) / [LinkedIn](https://www.linkedin.com/company/ton-blockchain)\n\n\n\n## **Background**\n\nTON was founded in 2018 as the “Telegram Open Network” by Telegram co-founders [Pavel Durov ](https://x.com/durov?lang=en)and [Nikolai Durov](https://x.com/Kolja_Durov?lang=en), and raised $1.7 billion across two private token sales in February–March 2018 to fund development. In October 2019, [the](https://aurum.law/newsroom/telegram-ton-1-7-bill-raise-sec-decentralization-the-legal-tale-and-insights) U.S. Securities and Exchange Commission (SEC) filed an emergency action against Telegram alleging an unregistered token offering, and Telegram ultimately ceased active involvement in May 2020. Development continued through the community-led “Newton” effort initiated by [Anatoliy Makosov](https://x.com/anatoly_makosov) and Kirill Emelyanenko, and in May 2021, the community voted to formalize governance under the TON Foundation and promote the V2 testnet into TON Mainnet.\n\nFollowing the community relaunch, TON attracted additional strategic and private investment. Since 2022, TON has [completed](https://messari.io/project/the-open-network/fundraising/funding) eight additional funding rounds, with at least three publicly disclosed raises totaling $50 million ($10 million from [DWF Labs](https://messari.io/organization/dwf-labs) in 2022, $30 million in a private sale backed by Foresight Ventures and Bitget in 2024, and $10 million in a strategic investment from [Gate](https://messari.io/organization/gate-io) in 2024).\n\nSince 2022, TON has attracted significant institutional backing. In [March 2025](https://www.tradingview.com/news/cointelegraph:d27a07a06094b:0-venture-capital-firms-invest-400m-in-ton-blockchain/), the TON Foundation disclosed that a group of investors, including Sequoia Capital, Pantera Capital, and Ribbit Capital, collectively purchased over $400 million in Toncoin. Combined with earlier rounds and subsequent investments from Coinbase Ventures, Pantera Capital, and others, publicly confirmed capital inflows exceed $550 million. Later in [August 2025](https://cryptoslate.com/verb-secures-558m-to-become-first-ton-treasury-vehicle-plans-ton-strategy-rebrand/), two publicly traded Toncoin treasury vehicles, [TON Strategy Co.](https://cryptoslate.com/verb-secures-558m-to-become-first-ton-treasury-vehicle-plans-ton-strategy-rebrand/) and [AlphaTON Capital](https://decrypt.co/news-explorer?pinned=1150165&title=portage-biotech-rebrands-as-alphaton-capital-corp-invests-100m-in-ton-digital-asset-treasury), have also launched, modeled on MicroStrategy's BTC accumulation strategy.\n\nIn [March 2025](https://www.tradingview.com/news/cointelegraph:d27a07a06094b:0-venture-capital-firms-invest-400m-in-ton-blockchain/), TON Foundation disclosed that a group of investors - including Sequoia Capital, Ribbit Capital, Benchmark, Draper Associates, Kingsway Capital, Vy Capital, Libertus Capital, CoinFund, SkyBridge, Hypersphere, and Karatage - collectively purchased and held $400 million in Toncoin. Pantera Capital, which has called Toncoin its largest single investment to date, reportedly deployed over $100 million. In [August 2025](https://www.theblock.co/post/366414/coinbase-ventures-buys-ton-tokens-directly-from-telegram-says-ton-foundation-exec), Coinbase Ventures purchased Toncoin directly from Telegram as a long-term venture position. In total, publicly confirmed capital inflows into TON exceed $550 million since 2022 across at least ten discrete investment events.\n\nA later turning point came through renewed alignment with Telegram distribution. In September 2023, the TON Foundation [announced](https://wublock.substack.com/p/ton-ecosystem-overview-deep-integration) a partnership with Telegram, aiming to integrate and promote the TON ecosystem for Telegram’s userbase. Thanks to this partnership, Telegram natively integrated a crypto wallet known as @Wallet, supported TON Sites in an in-app browser, and issued Fragment collections (usernames, collectable numbers, and gifts) on TON.\n\nA later turning point came through renewed alignment with Telegram distribution. In January 2025, TON [became](https://www.prnewswire.com/news-releases/ton-foundation-expands-partnership-with-telegram-as-ton-becomes-the-exclusive-blockchain-for-telegrams-mini-app-platform-302356251.html#:~:text=Jan%2021%2C%202025%2C%2010:,950%20million%20monthly%20active%20users.) the exclusive blockchain infrastructure for Telegram’s Mini App platform, with TON Connect positioned as the standard wallet connection method and Toncoin (TON) serving as the exclusive cryptocurrency for non-fiat payments across Telegram services (e.g., Stars, Premium, Ads). This shifted TON to a technically differentiated L1 focused on consumer distribution, with Telegram serving as its primary onboarding and application surface.\n\nToday, TON is maintained through a foundation and open-source contributors, so leadership is best understood through its public-facing executives and operating entities rather than a single corporate team. Makosov and Emelyanenko are cited as founding/core members of the TON Foundation. [Steve Yun](https://www.linkedin.com/in/steve-yun/) served as President of the TON Foundation Council and later launched the $100 million ecosystem venture fund, [TVM Ventures](https://www.linkedin.com/company/tvm-ventures/), in February 2025, while remaining on the board. [Maximilian Crown](https://www.linkedin.com/in/maxcrown/) was appointed CEO in April 2025 and later President in August 2025, with prior experience as a [MoonPay](https://www.moonpay.com/) co-founder (CFO/COO).\n\n\n\n## **Technology**\n\n\n\nTON’s architecture is commonly described as a “blockchain of blockchains.” Instead of relying on a single monolithic chain, TON is built as a hierarchical system in which a top-level chain coordinates with multiple parallel chains, which can be further subdivided. This structure is designed to enable scalability at the consumer scale while maintaining interoperability and shared security. At a high level, TON consists of three core components: (i) the masterchain, (ii) the workchains, and (iii) the shardchains, which are explained below in more detail.\n\n### **Architecture**\n\n#### **Masterchain**\n\nThe masterchain is the top-level coordinating chain. It does not process regular user transactions. Instead, it maintains critical network state, including (i) network configuration parameters, (ii) the validator set and their stakes, and (iii) references to the latest finalized blocks of all workchains and shardchains. In simple terms, the masterchain acts as the global source of truth for the TON network, ensuring consistency and finality across all parallel chains.\n\n#### **Workchains**\n\nWorkchains operate in parallel under the coordination of the masterchain. Each workchain can, in theory, define its own rules (e.g., virtual machine, token standards, or execution logic) while remaining interoperable within the TON ecosystem. The architecture theoretically supports up to 2³² workchains, though in practice, only the base workchain is currently active in most production contexts. The workchain layer enables long-term flexibility, allowing TON to support specialized environments without fragmenting security.\n\n#### **Shardchains**\n\nFrom there, each workchain can be split into shardchains, i.e., smaller partitions of the network state. Sharding allows transactions and smart contract execution to be processed in parallel across multiple shards rather than sequentially on a single chain. TON’s design theoretically supports up to 2⁶⁰ shardchains per workchain, although the network dynamically creates and merges shards as needed. The defining feature of TON’s scalability model is dynamic sharding. Rather than permanently fixing the number of shards, shardchains split automatically under high load to increase processing capacity and merge when activity declines, reducing unnecessary overhead. This allows network capacity to expand or contract in response to real-time demand. The goal is to maintain stable performance and predictable transaction times even as user activity scales to consumer levels.\n\n#### **Consensus**\n\nTON uses a Proof-of-Stake (PoS) consensus model. Validators are selected based on the amount of TON staked and participate in block production and validation. Consensus is achieved using a Byzantine Fault Tolerant (BFT) protocol called Catchain, which is specifically designed to operate efficiently in a sharded environment. Catchain enables validators to coordinate across shards while preserving security and finality guarantees.\n\n\n\nTON Core is [releasing Catchain 2.0](https://t.me/toncore/99), a consensus upgrade targeting sub-second finality to bring the onchain experience closer to Web2 responsiveness. The upgrade reduces block intervals from ~2.5s to 200–400ms, delivers roughly 2.5–5x throughput improvement, and cuts finalization lag from ~10s to ~1s.\n\nDevelopment is complete: the testnet, updated on January 23, 2026, has been running stably at ~450ms block intervals with ~1–2s finalization. Mainnet validators were updated on February 12, 2026 with the Catchain 2.0 code and accelerated network layer, though activation remains dormant pending final testnet validation.\n\n### **Validators**\n\nWithin the above architecture, there are two primary roles: (i) Validator and (ii) Nominators.\n\nValidators [secure](https://ton.org/en/validators) TON’s PoS network by staking Toncoin to participate in block production and validating transactions across shardchains. Operators must run high-performance, highly available infrastructure and stake a minimum of 300,000 TON to enter validator elections, though winning typically requires ~700,000 TON or more, depending on competition and the cycle's validator cap. Validators stake for a fixed validation term, with stake and rewards returned after the round completes.\n\nValidators [earn rewards](https://docs.ton.org/v3/documentation/nodes/validation/staking-incentives) from (i) transaction fee surpluses (users attach small extra Toncoin amounts as validator incentives) and (ii) newly issued Toncoin with proportional distribution based on stake weight. Validators can be [penalized](https://docs.ton.org/v3/documentation/smart-contracts/contracts-specs/governance) in two ways: (i) idle behavior: If a validator fails to participate in block creation or transaction signing during a validation round, it may be fined; and (ii) malicious misbehavior, where any network participant can submit a complaint with cryptographic proof to the [Elector contract](https://docs.ton.org/v3/documentation/smart-contracts/contracts-specs/governance). Validators vote on the complaint, and if 66.0% of validators approve, a slashing penalty is deducted from the validator's stake. To receive rewards, a validator must successfully (i) win election into a validation cycle, and (ii) validate blocks throughout the entire cycle without being penalized.\n\n[Nominators](https://docs.ton.org/v3/documentation/smart-contracts/contracts-specs/nominator-pool), on the other hand, (i) delegate TON to validators, and (ii) share in staking rewards. They do not operate the infrastructure directly. The cited minimum delegation amount is ~10,000 TON. Nominators increase validator stake weight and help decentralize participation in consensus.\n\nBeyond direct nomination, [pooled staking](https://docs.ton.org/ecosystem/staking/overview) services such as TON Whales, Kiln, ChorusOne, and P2P aggregate smaller stakes to meet validator thresholds. Alternatively, [liquid staking](https://docs.ton.org/ecosystem/staking/overview) protocols (Tonstakers, KTON, bemo, Hipo) issue transferable receipt tokens that let holders earn staking rewards while keeping capital liquid across DeFi.\n\n### **AI**\n\nTelegram is emerging as a native interface for AI agents, with TON positioning itself as the settlement and infrastructure layer for this convergence.\n\n[Cocoon](https://cocoon.org/) (Confidential Compute Open Network) is a decentralized AI compute network built on TON, [announced](https://x.com/durov/status/1995208789600182391?lang=en) by Pavel Durov at Blockchain Life 2025 and launched on mainnet in December 2025. GPU owners contribute computing power and earn Toncoin, while user data remains encrypted throughout execution via Trusted Execution Environments (TEEs). Telegram already routes lightweight AI operations through Cocoon, including message translation and summarization, with heavier workloads such as media processing and conversational assistants planned next. With Telegram's 1B+ user base as built-in distribution and Telegram’s recent efforts to simplify the creation of [agentic products](https://telegram.org/blog/ai-editor-mighty-polls-and-more), Cocoon represents one of the most significant real-world deployments of decentralized AI compute to date.\n\nTelegram's bot infrastructure has also evolved toward AI-native use cases. In early 2026, Telegram shipped [streaming responses for bots](https://t.me/BotNews/113) (purpose-built for AI assistants) and threaded conversations for multi-topic chats. Third-party traction reinforces the direction: OpenClaw, the most popular self-hosted AI assistant framework, defaults to Telegram as its messaging layer. Additionally, a [Telegram developer](https://x.com/steipete/status/2037197024081195188) became its new maintainer. [BotFather](https://identityhub.app/blog/telegram-default-ai-interface), Telegram's built-in tool for creating and configuring bots, reached 7.3M monthly active users (MAUs) by early 2026, more than doubling from 3.5 million in January 2025. Combined with native Toncoin withdrawal support for bot earnings, these updates tighten the economic link between Telegram's bot ecosystem and the TON network.\n\nTON's deeper integration with onchain AI agents remains at an earlier stage. To accelerate development, TON Foundation is running dedicated [AI contests](https://identityhub.app/contests) to bootstrap tooling and adoption.\n\n## **Putting It All Together**\n\nThe components described above (i.e., masterchain coordination, dynamic shardchains, asynchronous smart contracts, TVM execution, cell-based data storage, and hypercube routing) operate within a broader multi-layer architecture designed for consumer-scale usage inside Telegram. Rather than functioning as a single monolithic execution engine, TON operates as an interconnected system composed of: (i) User Interface Layer, (ii) Application Layer, (iii) Execution Layer, (iv) Routing & Sharding Layer, (v) Consensus Layer, (vi) Validator Infrastructure Layer, (vii) Indexing & API Layer, (viii) Storage & Data Layer. Together, these layers allow TON to support real-time financial interactions embedded directly inside Telegram’s social graph.\n\nFor example, consider a user sending USDT to a friend inside Telegram:\n\n* The transaction begins inside the Wallet in Telegram. The user enters an amount and taps send. The experience feels identical to sending a message.\n* The wallet constructs a Jetton (TEP-74) transfer transaction. Because TON uses an asynchronous model, a message is sent to the sender’s USDT contract, which then sends a follow-up message to the recipient’s wallet contract.\n* The TON Virtual Machine executes the smart contract logic:\n * Validates balances\n * Deducts TON for gas\n * Updates Jetton balances\n * Emits internal messages\n * Each contract runs independently\n* If the sender and receiver reside on different shardchains, the message is routed using Hypercube Routing. Rather than broadcasting globally, the system calculates the shortest path between shards, minimizing latency.\n* If network load increases, shardchains may automatically split to distribute execution load. This is TON’s “infinite sharding” in action.\n* Validators on the relevant shardchain produce and validate the block. The masterchain later finalizes shard references to ensure global consistency.\n* Collators assemble transactions while validators confirm them, enabling parallelization and improved stability.\n* Toncenter APIs and indexers immediately update transaction status. Wallets can display “pending” and then a confirmed state using trace APIs and action parsing.\n* All state changes are stored in TON’s cell-based structure and packaged into a Bag-of-Cells (BoC), ensuring compact storage and verifiable hash-linked data.\n\nTo the user, this entire multichain, asynchronous, routed, validated process appears as a simple chat-based transfer.\n\n\n\n## **Toncoin (TON Token)**\n\n### **Token Functions**\n\nAs the project’s documentation outlines, Toncoin is a native coin of the TON blockchain. It serves several key functions within the network, including:\n\n* **Paying for network execution:** Toncoin is required to execute transactions (e.g., asset transfers and swaps), with fees paid in Toncoin and designed to remain low for consumer-scale usage.\n* **Securing the network:** Validators stake Toncoin to participate in PoS consensus and earn rewards, with penalties for downtime or misbehavior. Nominators can delegate Toncoin to share in validator staking rewards.\n* **Telegram in-app economy:** Telegram uses TON as its blockchain layer for ownership, transfers, and payouts. Toncoin serves as the payment rail for collectible usernames, SIM-less phone numbers, and collectible gifts - all of which are onchain assets. Creator earnings from ads and in-app purchases (for channel authors, bot developers, and mini-app owners) are withdrawn exclusively via **Fragment.com** in Toncoin. Toncoin is also the currency for purchasing ads on the **Telegram Ads** platform.\n* **Storage of blockchain data:** Toncoin is required to maintain smart contracts onchain, where validators charge storage rent proportional to the data held. For larger files, TON Storage provides a decentralized persistence layer where storage providers are paid in Toncoin and must cryptographically prove file integrity to claim rewards.\n\n### **Tokenomics**\n\n\n\nThe initial TON supply of 5 billion tokens was placed into [20](https://ton.org/en/mining) Proof-of-Work Giver smart contracts and mined permissionlessly between July 2020 and June 2022. Since the PoS transition (June 28, 2022), new TON is minted via validator block rewards (~0.5–0.7% annually). The current total supply is ~5.16 billion TON. At [$1.33](https://www.coingecko.com/en/coins/toncoin) (Mar. 26, 2026), this implies a fully diluted valuation (FDV) of ~$6.9 billion. Note: TON has no max supply cap.\n\nThe pie chart above depicts the current supply distribution of ~5.15 billion Toncoin as of March 2026, broken into 14 categories. For simplicity, these can be grouped into four macro buckets:\n\n* **Freely Circulating:** ~48% (~2.46B TON). Includes Regular Wallets, Staking (Elector), Centralized Exchanges, CEX Custodial, DeFi, Smart Contracts & Others, Uninit Wallets, TON Ecosystem Reserve, TON Foundation, and Other Labeled.\n* **TON Believers Fund:** 25% (~1.32B TON). A voluntary lockup where existing holders deposited tokens into a [Locker smart contract](https://github.com/ton-blockchain/locker-contract) (Jul–Oct 2023). Now vesting in 36 monthly installments (~36.6M TON/month) through Oct 2028. As of March 2026, 6 of 36 periods have been completed; actual claim rates have been low, with only a fraction of unlocked tokens withdrawn so far. [Onchain data.](https://tonviewer.com/UQDtFpEwcFAEcRe5mLVh2N6C0x-_hJEM7W61_JLnSF74p9dz)\n* **Frozen Inactive Miners:** 20.9% (~1.08B TON). 171 addresses that mined TON during the PoW phase but never transacted. [Frozen by community governance vote](https://blockworks.co/news/ton-governance-votes-to-freeze-20-of-its-ton-supply) (Feb 2023) for 48 months. After expiration (~Feb 2027), owners must manually activate, and many keys are likely lost. The community could vote to extend the freeze or burn these tokens before expiration, though no formal proposal has surfaced as of this writing. [Address list.](https://tontech.io/stats/#/early-miners)\n* **Telegram:** ~6% (~327M TON). Tokens held by Telegram, plus ~21M in vesting contracts deployed to team members and partners (1,440-day vesting, 360-day cliff). Sell pressure from vesting recipients has been negligible thus far.\n\n### **Token Vesting**\n\n[Unlock schedules](https://defillama.com/unlocks/ton) for the parties mentioned above include:\n\n* **IPoW Mining (Premine):** Fully mined and distributed. The original ~5B TON was mined from [20 Giver contracts](https://ton.org/en/mining) between July 6, 2020, and June 28, 2022. Mining was permissionless with a variable rate based on PoW difficulty, not a linear release schedule. No further unlocks.\n* **TON Believers Fund:** Unlocks through 36 fixed monthly releases of ~36.59 million TON/month (~2.8% of the fund). As of March 2026, 6 of 36 periods have been completed; 30 periods remain (~1.098 billion TON still to unlock through ~Oct. 2028).\n* **Frozen Inactive Miners:** The 48-month [freeze](https://blockworks.co/news/ton-governance-votes-to-freeze-20-of-its-ton-supply) on 171 addresses (~1.081 billion TON) expires around Feb. 2027. This does not mean tokens enter circulation: these wallets have never transacted, owners must manually activate them, and many keys are likely permanently lost. The TON community could also vote to extend the freeze or burn these tokens before expiration. [Address list.](https://tontech.io/stats/#/early-miners)\n* **[Vesting contracts](https://github.com/ton-blockchain/vesting-contract):** Use a 1,440-day schedule with a 360-day cliff. Most early contracts have already fully vested; remaining contracts continue vesting at ~3.45M TON/month (per CEX listing model).\n* **PoS Emission (Inflation):** Ongoing, perpetual issuance at[ ~73,000–97,000 TON/day](https://www.tonstat.com/) (1.7 TON per masterchain block + 1.0 TON per basechain block). 50% of transaction fees are [burned](https://dune.com/ton_foundation/staking).\n\n### **Governance**\n\nTON's governance spans three layers: validators, core development, and ecosystem coordination.\n\nAt the protocol layer, approximately 400 validators distributed across 40 countries, with over 450M Toncoin staked, govern upgrade decisions directly. Any change to network parameters or consensus rules must pass an onchain vote among active validators, ensuring no single entity can push through protocol changes unilaterally. According to Chainspect, TON ranks 3rd among Layer-1s by Nakamoto coefficient, placing it among the most decentralized Proof-of-Stake networks.\n\nTON Core, the network's core development arm, maintains the TON node software, ships protocol upgrades, and builds node tooling. The TON Foundation, a non-profit dedicated to the ecosystem's long-term growth, provides grants, resources, and technical support to projects building on TON.\n\nBeyond these two bodies, a growing set of independent teams contributes to infrastructure and developer tooling. TonTech, an engineering team supported by the TON Foundation, maintains core developer primitives including AppKit, AgentKit, WalletKit, and TON Connect. RSquad, a blockchain development team active in the TON ecosystem since its early days, has contributed critical infrastructure, including the Rust TON Node, TON Pay, and TON Teleport, a trustless cross-chain bridge for asset transfers between TON and external networks.\n\n## **TON Ecosystem**\n\n\n\n### **Partners and Projects**\n\nKey projects that highlight the variety of benefits the TON ecosystem and blockchain provide:\n\n* **[The Open Platform (TOP)](https://top.co/):** The largest Web3 product development company within the Telegram ecosystem, building and investing in infrastructure and consumer applications on TON. TOP's portfolio includes Wallet in Telegram, Tonkeeper, STON.fi, and Getgems, and the company reached a [$1 billion valuation](https://www.theblock.co/press-releases/361030/the-open-platform-is-first-unicorn-in-web3-ecosystem-in-telegram-at-1bn-valuation) in 2025 after raising over $70 million from Ribbit Capital, Pantera Capital, and others.\n* **[Wallet in Telegram](https://wallet.tg/ton):** A crypto wallet natively integrated into Telegram, developed by TOP, supporting both custodial and self-custodial modes dependent on region. It serves as the primary onramp for Telegram's 1B+ user base, enabling in-chat transfers, Toncoin purchases, and direct access to Telegram Mini Apps without leaving the messenger.\n* **[Tether (USDT)](https://tether.to/):** The dominant stablecoin on TON by circulating supply and the default asset for payments and most DeFi activity across the ecosystem. TON currently holds roughly [$1.28 billion in stablecoins](https://defillama.com/stablecoin/tether), with USDT serving as the primary settlement unit for wallets, DEXs, and merchant payments.\n* **[Ethena](https://ethena.fi/):** Introduced synthetic yield-bearing digital dollars (USDe and sUSDe) into TON's Telegram-native DeFi stack, expanding stablecoin use cases beyond payments into passive yield products. Eligible users holding [tsUSDe](https://blog.ton.org/ethena-ton-foundation-usde-on-ton) in a TON wallet earn boosted yields, with plans for neobanking and peer-to-peer payments powered by Ethena within Telegram.\n* **[xStocks](https://www.ton.org/en/x-stocks-are-live-on-ton-real-world-stocks-now-on-chain):** Tokenized U.S. equities launched on TON, bringing real-world stock exposure (e.g., Apple, Tesla, Microsoft) directly into TON wallets with a self-custodial UX. The platform currently supports over 60 tokenized stocks and ETFs powered by [Backed Finance](https://www.coindesk.com/business/2026/03/10/kraken-s-tokenized-stock-venue-starts-points-program-hinting-at-possible-ecosystem-token/) under Kraken's institutional framework, with plans to expand to 500+ by end of 2026.\n* **[Fragment](https://fragment.com/):** Onchain marketplace integrated into Telegram where collectible usernames, SIM-less phone numbers, and digital gifts are minted and traded as NFTs powered by Toncoin. Fragment is the primary driver of TON's [#2 ranking in NFT trading volume](https://dune.com/ton_foundation/nft) behind Ethereum, with transaction activity tied directly to Telegram's social graph.\n\nOther notable projects building on TON include [Affluent](https://www.affluent.org/), [Bidask](https://bidask.finance/), [Boinkers](https://t.me/boinker_bot), [CapsGame](https://t.me/capsgamebot), [DeDust](https://dedust.io/), [Gamee](https://t.me/gamee), [Gatto](https://t.me/gattoton_bot), [Getgems](https://getgems.io/), [MyTonWallet](https://mytonwallet.app/), [RedStone](https://redstone.finance/), [Storm Trade](https://stormtrade.io/), [Swap Coffee](https://swap.coffee/), [TeleTON](https://teletonagent.dev/), [TONCO](https://tonco.io/), [Torch Finance](https://torch.finance/), [xRocket](https://t.me/xrocket_bot), and [Zargates](https://zargates.com/).\n\nRecent partnerships and integrations:\n\n* [**Mar. 31, 2026**](https://www.tradingview.com/news/cointelegraph:11581f704094b:0-dynamic-adds-embedded-wallet-infrastructure-to-ton-for-telegram-apps/): [Dynamic](http://www.dynamic.xyz) launched embedded wallet infrastructure for TON, letting developers automatically deploy wallets inside their apps and Telegram Mini Apps.\n* **[Mar. 26, 2026](https://blockchain.news/flashnews/walletconnect-integrates-with-ton-blockchain-for-seamless-connectivity):** [WalletConnect](https://walletconnect.com/) launched production support on TON, enabling standardized wallet connections across dApps and Telegram Mini Apps.\n* **[Feb. 17, 2026](https://www.theblock.co/post/390138/ton-foundation-osl-banxa-stablecoin-payments):** [TON Foundation](https://ton.foundation/en) partnered with OSL's Banxa to expand stablecoin payment infrastructure for Asia-Pacific merchants.\n* **[Jan. 7, 2026](https://x.com/ton_blockchain/status/2008966919530090821):** TON announced Toncoin support in [Atomic Wallet](https://atomicwallet.io/), expanding TON's reach into a multichain, non-custodial wallet user base.\n* **[Dec. 24, 2025](https://x.com/ton_blockchain/status/2003764859138199985):** [Fonbnk](https://www.fonbnk.com/) + [Tether](https://tether.to/) expanded USDt on Telegram Wallet, positioning mobile-money conversion as the core bridge for Africa-focused remittances and stablecoin banking flows.\n* **[Dec. 23, 2025](https://x.com/ton_blockchain/status/2003490653645246825):** [HoudiniSwap](https://houdiniswap.com/) launched private TON payments, allowing users to request payments without exposing wallet addresses or transaction history, with inbound support across 120+ chains.\n* **[Dec. 22, 2025](https://investors.shift4.com/news-events/press-releases/detail/288/shift4-launches-global-stablecoin-settlement-platform-unlocking-faster-payments-for-merchants):** [Shift4](https://www.shift4.com/) launched a global stablecoin settlement platform, unlocking faster payments for merchants.\n* **[Aug. 5, 2025](https://x.com/ton_blockchain/status/1952723911130456184):** [Zengo](https://zengo.com/) wallet added native Toncoin support, expanding TON's presence across non-custodial mobile wallets.\n* **Upcoming:** [Chainlink CCIP](https://chain.link/cross-chain) cross-chain interoperability integration (announced 2025, launch pending). [Revolut](https://www.revolut.com/) soft-launched TON support in select regions, with broader availability now live.\n* Toncoin is also listed on major U.S. exchanges, including [Robinhood](https://thedefiant.io/news/nfts-and-web3/robinhood-adds-toncoin-to-u-s-crypto-platform-ahead-coinbase-563a1ec6) (Aug. 2025), [Gemini](https://www.gemini.com/blog/toncoin-usdton-is-now-available-on-gemini) (Sep. 2025), and [Coinbase](https://www.businesswire.com/news/home/20251118274287/en/TON-Strategy-Company-Welcomes-Coinbases-Launch-of-$TON-Trading-Across-Global-Platforms) (Nov. 2025).\n\nFor a recap of category-specific ecosystem developments, please read the following [DeFi](https://blog.ton.org/defi-on-ton), [Institutional](https://blog.ton.org/institutions-on-ton), and [NFT](https://blog.ton.org/how-nfts-evolved-on-ton) reports.\n\n### **Network Metrics**\n\n\n\nTON’s network activity in 2025 reflects a chain that has moved past the initial hype spike and is settling into a more durable, consumer-driven usage baseline. Daily active users peaked sharply in early 2025 (~600K), then normalized throughout the year, finishing Q4 2025 with ~1.0% QoQ growth and a relatively stable range of ~100K–150K.\n\n\n\nIn parallel, TON has sustained meaningful throughput, with daily transactions spiking above ~7 million during early 2025 surges, then stabilizing around ~1.5–2.5 million/day, with periodic bursts and an end-of-year lift similar to daily user data.\n\nTON's NFT market ranks [second only to Ethereum by trading volume](https://dune.com/ton_foundation/nft), driven largely by Telegram-native assets like collectible usernames, numbers, and gifts.\n\n## **Roadmap**\n\nOn Jan. 22, 2025, the TON Core team [published](https://t.me/toncore/6) its H1 2025 roadmap, centered on shipping the long-in-the-works “Accelerator” upgrade, an architecture-level refactor intended to better realize the sharded execution model described in the TON whitepaper and to keep performance stable as load scales. The roadmap prioritized (i) scaling and stability at the protocol layer, (ii) validator operability and resilience, and (iii) faster, more human-readable UX through better APIs and indexing.\n\nBuilding on the Accelerator foundation, TON's [2026 roadmap](https://ton.org/en/roadmap) shifts focus from the infrastructure-level refactoring of the Accelerator era toward developer accessibility and its top priority, sub-second finality. At the protocol layer, Catchain 2.0 targets sub-second block finalization, while the Rust Node brings institutional-grade operability and resilience to the validator set.\n\nOn the developer side, the roadmap prioritizes four workstreams: (i) smart contract tooling through [Tolk 1.3](https://docs.ton.org/v3/documentation/smart-contracts/tolk/overview) and toolchain, the successor language and SDK stack to FunC, offering TypeScript/Rust-inspired syntax and up to 40% lower gas costs; (ii) [AppKit](https://ton.org/dev/appkit), a unified application layer via development kits for apps, wallets, and payments ([TON Pay](https://ton.org/en/ton-pay-a-new-payments-layer)), compressing the path from idea to shipped Telegram Mini App; (iii) vibe-coding workflows that let developers describe an app to an AI agent and receive a working prototype, ready to share on Telegram; and (iv) AgentKit, an MCP-based toolkit giving autonomous AI agents structured access to wallets, transfers, and DeFi modules on TON.\n\n## **Closing Summary**\n\nTON is positioning itself as one of the few L1s explicitly engineered for consumer-scale adoption, not just DeFi-native throughput benchmarks. Its core bet is that mainstream crypto use breaks on two constraints, distribution and latency/cost, and that solving both requires more than a fast chain. TON’s dynamically sharded, asynchronous architecture is designed to maintain stable performance under load, while its vertically integrated protocol services (DNS, Storage, Payments, Proxy, Sites) reduce reliance on third-party infrastructure. Most importantly, TON’s deep integration with Telegram collapses the traditional crypto UX stack (wallet → browser → dApp) into a single messaging-native environment where onchain actions can feel like normal in-app behavior.\n\nIn 2025, TON shifted to a technically differentiated L1 focused on consumer distribution, with Telegram serving as its primary onboarding and application surface. The ecosystem has increasingly matured into a Telegram-native financial stack while network activity reflects a chain transitioning from hype-driven spikes to a steadier transactional baseline.\n\nLooking forward, TON's 2026 roadmap shifts from infrastructure refactoring to performance and developer accessibility. Catchain 2.0 targets sub-second finality, the Rust Node reimplements the validator stack, and a unified developer layer (Tolk, AppKit, TON Pay) compresses the path from idea to shipped Telegram Mini App. In parallel, Telegram is becoming a native interface for AI agents, with Cocoon providing decentralized compute and AgentKit connecting autonomous agents to onchain actions. If the vibe-coding loop works at scale, where builders generate working prototypes from a natural-language prompt, share them inside Telegram for instant feedback, and iterate daily rather than quarterly, TON becomes not just a settlement layer for digital finance inside Telegram, but the fastest path from idea to testable product in crypto.",
"hook": "TON has quietly assembled the ingredients most Layer-1s lack: 1B+ captive users via Telegram, $550M+ in institutional capital, and a financial stack (stablecoins, tokenized equities, yield products) embedded directly in the messenger. This IoC covers the architecture, tokenomics, ecosystem, and 2026 roadmap driving TON's bet that consumer crypto adoption starts inside the chat window.",
"publishDate": "2026-04-07T16:00:00Z",
"readingTimeInMinutes": 21.80333333333333,
"slug": "understanding-ton-a-comprehensive-overview",
"subscriptionTier": "free",
"summary": "TON is the only Layer-1 with native distribution inside a 950M MAU messaging app. This Initiation of Coverage breaks down the network's architecture, validator economics, DeFi ecosystem, tokenomics, and 2026 roadmap. At ~$6.9B FDV, the question is whether Telegram's user base converts into sustained onchain activity.",
"tags": [
{
"id": "8ef53cf6-d66a-48e3-bf0d-03da8cfdca14",
"name": "Consumer"
},
{
"id": "9d1f93f6-d15c-40ea-b384-58c2a9419fd8",
"name": "Layer-1"
}
],
"title": "Understanding TON: A Comprehensive Overview",
"previewImage": "https://cdn.sanity.io/images/2bt0j8lu/production/cf321d93bc0b99464d29d02c79b4eda23941bfeb-6400x3600.png"
},
{
"id": "ce62bc98-3825-479a-b419-18b33d4ad5f0",
"createdAt": "2026-04-02T20:16:42Z",
"updatedAt": "2026-04-02T20:44:02Z",
"assetIds": [
"9cc76ef5-f68f-45b3-930d-98af8f6513fc"
],
"assets": [
{
"id": "9cc76ef5-f68f-45b3-930d-98af8f6513fc",
"name": "Sai",
"symbol": "Sai",
"slug": "sai-dot-fun"
}
],
"authors": [
{
"id": "cbe1f170-343a-461c-abd3-9ceee66022b8",
"name": "Eric Manoukian",
"image": "https://cdn.sanity.io/images/2bt0j8lu/production/d311e4315bfaa623b347b8ad6d1ce1e4f48023ff-400x400.jpg?w=100",
"linkedinUrl": ""
}
],
"content": "## Key Insights\n\n* Sai is a perpetual decentralized exchange (perp DEX) built on Nibiru that launched publicly on Feb. 18, 2026. It offers up to 100x leverage with oracle-settled pricing and pools liquidity through Sai Liquidity Positions (SLPs) that back markets and absorb trader P&L.\n* The launch coincided with Let's Go Saicho, a $25,500 trading competition split into two phases. Phase 1 (Feb. 18 to March 4) rewarded the top 25 traders by percentage ROI with up to $20,000, and Phase 2 (March 5 to March 19) distributes $5,500 across volume-based tracks.\n* As of March 16, TVL reached $47,211, cumulative perp volume hit $6 million, and the platform collected $21,560 in fees. Open interest peaked at $549,628 on March 3, the same day volume reached its single-day high of $1.27 million.\n* Sai saw 134 unique traders, with 52.9% executing six or more trades. Returning traders accounted for 58.2% of Phase 2 daily activity. The liquidation rate dropped from 12.9% in Phase 1 to 6.7% in Phase 2 despite average leverage increasing from 33.1x to 44.2x.\n* The roadmap extends beyond perps: a white-label Perps-as-a-Service product, Sai Savings with ~5% yield on idle stables, automated strategy vaults, real-world asset markets, and a mobile application positioning Sai as trading infrastructure rather than an isolated venue.\n\n## Primer\n\n[Sai](https://sai.fun/) is a perpetual decentralized exchange (perp DEX) that aims to simplify, predict, and make leveraged trading more accessible. Users [connect](https://docs.sai.fun/guides/wallet-setup) to the application with standard EVM wallets, deposit collateral such as USDC or stNIBI, and trade perp markets through an interface that resembles a centralized futures exchange. They can open long or short positions, set up to 100x leverage on certain assets, and use familiar order types, including market, limit, stop, and conditional (stop-loss/take-profit) orders.\n\nSai was created by members of the Nibiru team with experience in distributed systems, infrastructure, and financial protocols. The creators are led by Nibiru founder and CEO [Unique Divine](https://www.linkedin.com/in/unique-divine/), who has a background in applied mathematics and machine learning, as well as prior experience at [IBM](https://www.ibm.com/us-en) and [Sommelier Protocol](https://somm.finance/).\n\nSai is built on [Nibiru](https://nibiru.fi/) Chain, a high-throughput Layer-1 with a unified EVM and [Wasm execution](https://nibiru.fi/docs/concepts/wasm/) environment. Nibiru combines an EVM-equivalent runtime ([Nibiru EVM](https://nibiru.fi/docs/evm/)) and a Wasm environment in a single state machine, allowing Solidity and Rust contracts to coexist, share accounts and gas, and call each other through built-in system contracts that bridge the two VMs. Its consensus layer, [Nibiru BFT](https://nibiru.fi/docs/concepts/arch/nibiru-bft/), is an evolution of [CometBFT](https://docs.cometbft.com/v0.38/), delivering fast finality and sub-two-second settlements. Alongside core modules for staking, governance, IBC, and a native oracle, this architecture provides Sai with the low-latency execution, deterministic settlement, and oracle support necessary to offer a CEX-like onchain derivatives platform. Sai’s defining characteristic is its approach to pricing and risk: execution is anchored to decentralized oracles and backed by pooled vaults called Sai Liquidity Positions (SLPs), emphasizing consistent behavior across different market regimes.\n\n## Let’s Go Saicho: The Trading Competition\n\nSai's [public launch](https://x.com/SaiDotFun/status/2024080216474497407?s=20) on Feb. 18, 2026, coincided with the start of [Let's Go Saicho](https://docs.sai.fun/resources/blogs/lets-go-saicho), a one-month trading competition designed to bootstrap both trading activity and platform liquidity. The competition runs through March 19 and distributes a $25,500 prize pool across two phases, each targeting a different type of trader behavior.\n\n### Phase 1: PnL Competition\n\n[Phase 1](https://docs.sai.fun/resources/blogs/lets-go-saicho#phase-1-pnl-competition-feb-18-march-4) ran from Feb. 18, 2026, to March 4, 2026, and allocated $20,000 to the top 25 traders ranked by percentage ROI rather than absolute profit. A trader who earned a 50% return on a $500 account could outrank a trader who earned 5% on a $50,000 account. Only closed positions counted toward PnL, and unrealized gains were excluded.\n\nEligibility thresholds scaled with rank. Traders competing for the top three positions needed at least $1 million in cumulative volume and $250 in profit. Ranks 4 through 10 required $50,000 in volume and $50 in profit, while ranks 11 through 25 required $50,000 in volume with no minimum profit. The prize distribution was structured as follows:\n\n* Rank 1: $6,250\n* Rank 2: $3,125\n* Rank 3: $1,250\n* Rank 4-10: $625 each\n* Ranks 11-25: $250 each\n\n\n\nNo [traders](https://app.sai.fun/leaderboard/) met the $1 million volume threshold required for the top three positions, so none of the top three prizes were distributed. The remaining prizes were awarded to qualifying traders in ranks 4 through 25.\n\n### Phase 2: Volume Competition\n\n[Phase 2](https://docs.sai.fun/resources/blogs/lets-go-saicho#phase-2-volume-based-march-5-march-19-live-now) ran from March 5, 2026, to March 19, 2026, and shifted the incentive from profitability to volume, distributing $5,500 across three tracks. The largest pool, $4,000, is shared among all traders who cross $50,000 in volume, with each trader's share proportional to their total volume. All activity from Phase 1 counts toward Phase 2 thresholds in this segment. A $1,000 pool is split evenly among the first 50 traders to reach $10,000 in Phase 2 volume, creating an early-mover incentive. A single $500 prize goes to the highest-volume trader in Phase 2.\n\nUnlike Phase 1, Phase 2 did not require profitability to qualify. The competition enforces rules against sybil attacks, wash trading, fake volume, and malicious bots. A minimum position holding time of 10 to 20 minutes applies during Phase 2 to discourage instantaneous round-trip trades designed to inflate volume. Winners are expected to be announced shortly.\n\n## Early Traction and Key Metrics\n\n### Volume\n\n\n\nCumulative [perpetual volume](https://defillama.com/protocol/sai?tvl=false) from Feb. 18, 2026, through March 16, 2026, reached $6 million. Phase 1 accounted for $3.1 million of that total, with the final three days of Phase 1 generating $2.4 million as traders pushed to lock in PnL rankings. Phase 2 volume through March 16, 2026, totaled $2.9 million across 12 days, a higher daily average than Phase 1's first 10 days but below the Phase 1 closing sprint. Volume spiked again on March 11, 2026, at $573,420 before tapering to $36,810 on March 16, 2026, the lowest daily figure since launch, as Phase 2 approached its final days.\n\n### TVL and Open Interest\n\n\n\nAs of March 16, 2026, Sai's [total value locked](https://defillama.com/protocol/tvl/sai) (TVL) was at $47,210, up 14x from $3,380 on launch day. Growth came in two distinct waves. TVL held relatively steady between $3,200 and $4,400 during Sai's first week, then increased to $15,140 on Feb. 25, 2026, and climbed to $26,650 by Feb. 28, 2026, as early competition activity drew deposits. A second leg carried TVL from $29,120 on March 5, the start of Phase 2, to $47,210 by March 16, a 62% increase over 12 days, driven in part by new SLP deposits entering the platform after Phase 1 concluded.\n\n[Open interest](https://defillama.com/protocol/sai?openInterest=true) peaked at $549,630 on March 3, the same day perp volume hit its single-day high of $1.27 million. By March 16, open interest had settled to $27,620 as the competitive intensity of Phase 1's final days gave way to Phase 2's steadier volume-building pace.\n\n### Fees\n\n\n\nSai collected $21,560 in cumulative [fees](https://defillama.com/protocol/sai?tvl=false&perpVolume=false&fees=true) from launch through March 16. Phase 1 generated $19,600, and Phase 2 contributed $1,960 through its first 12 days. Daily fees peaked at $3,310 on Feb. 27, 2026, during a stretch from Feb. 24, 2026, to March 1, 2026, where fees exceeded $1,000 every day. After the Phase 2 transition, daily fees dropped sharply and remained subdued, falling below $100 on four of the first five days as trading shifted from high-conviction PnL plays to lower-cost volume accumulation.\n\nThe effective fee rate, fees as a percentage of volume, averaged 0.36% across the full period but diverged between phases at 0.64% in Phase 1 versus 0.07% in Phase 2. This gap likely reflects differences in sizing and trade mechanics. Phase 1 traders opened larger positions and held them through wider price moves to accumulate PnL, generating more fee-bearing events per dollar of volume. Phase 2 traders used smaller positions with higher leverage, producing high notional volume relative to the fees collected.\n\n### User Activity and Trade Frequency\n\n\n\nAs of March 16, 2026, Sai had 134 [unique traders](https://dune.com/queries/6825342/10709246). The platform had an existing base of 39 traders from its private launch period before Feb. 18, 2026, and the Let’s Go Saicho competition brought in 95 additional participants, 62 during Phase 1 and 33 during Phase 2.\n\n\n\nThe trader frequency [distribution](https://dune.com/queries/6825355/10709839) reveals a relatively engaged base for a new platform. Only 3.4% of traders executed 1 trade. The largest cohort, 43.6%, executed 2 to 5 trades, while 35.9% places 6 to 20 trades. The most active decile, 10 traders at 50+ trades each, averaged 11.8 active trading days, indicating a core group that traded consistently across both phases.\n\n### User Retention\n\n\n\n[Retention](https://dune.com/queries/6825351/10709830) patterns during Phase 2 provide a window into user stickiness. Returning traders [accounted](https://dune.com/queries/6825349/10709258) for 58.2% of daily active trader appearances from March 5, 2026, through March 16, 2026. After the March 5 spike of 60 active traders, including 37 new arrivals, daily activity settled at an average of 8.7 traders. The pattern suggests that Sai retained a meaningful share of Phase 1 participants in Phase 2 but struggled to attract new entrants after the Phase 2 launch-day momentum passed. \n\n### Risk Behavior: Liquidations and Leverage\n\nThe shift in competition incentives produced a clear change in risk behavior. The [liquidation rate](https://dune.com/queries/6825407/10709162) dropped from 12.9% of trades in Phase 1 to 6.7% in Phase 2, a 48% relative decline. This occurred despite [average leverage](https://dune.com/queries/6825407/10709162) increasing from 33.1x to 44.2x. Phase 1’s ROI-based rewards incentivized directional bets with higher risk tolerance, meaning getting liquidated was an acceptable cost to pursue outsized returns. Phase 2’s volume-based rewards penalized liquidations indirectly, since blown-up capital can’t generate more volume.\n\n## Community: The SaiClone Ambassador Program\n\nAlongside the trading competition, Sai launched the [SaiClone Ambassador Program](https://docs.sai.fun/resources/blogs/saiclone-ambassador), a three-tier progression system that operates entirely through Sai's [Discord](https://discord.com/invite/saidotfun) server. The program uses the [Mee6](https://mee6.xyz/en) bot to track contributions and assign XP, rewarding community engagement, content creation, and platform advocacy across the following tiers:\n\n* [Saicho](https://docs.sai.fun/resources/blogs/saiclone-ambassador#id-1.-saicho) (levels 0-5, up to 1,624 XP): The entry tier, granted automatically upon engaging with the Discord community.\n* [Saiborg](https://docs.sai.fun/resources/blogs/saiclone-ambassador#id-2.-saiborg) (Levels 6-15, 1,625-13,799 XP): Trusted, highly engaged members who receive special community recognition and increased influence on community decisions.\n* [Sage](https://docs.sai.fun/resources/blogs/saiclone-ambassador#id-3.-sage) (Level 16+, 13,800+ XP): Reserved for top contributors. Unlike other tiers, Sage requires a formal application reviewed by the Sai team. Benefits include direct team access, exclusive merchandise, bi-monthly raffles, collaboration opportunities, and early access to platform updates.\n\n[XP](https://docs.sai.fun/resources/blogs/saiclone-ambassador#how-to-earn-xp) accrues through trading activity, technical analysis and signals, community events, video and educational content, X engagement, bug reports, and Discord messages. The program ties community growth directly to platform usage, creating a feedback loop between trading activity and ambassador progression.\n\n## Looking Ahead\n\nSai's roadmap positions the platform as a trading infrastructure rather than a standalone venue. The most differentiated planned product is a white-label Perps-as-a-Service offering, with the first [iteration](https://x.com/SaiDotFun/status/2032458407899209822) from [Coded Estate](https://app.codedestate.com/perps/trades), an RWA platform that now offers perpetual trading powered by Sai. If executed, this would extend Sai's liquidity and infrastructure beyond its own front end.\n\nOn the product side, planned additions include [Sai Savings](https://docs.sai.fun/resources/blogs/intro-to-sai#new-products-and-yield:~:text=products%20and%20yield-,Sai%20Savings,-%3A%20A%20way%20for), which targets ~5% yield on idle stable balances, [automated strategy vaults](https://docs.sai.fun/resources/blogs/intro-to-sai#new-products-and-yield:~:text=control%20and%20flexibility-,Automated%20strategies,-%3A%20Launch%20vaults), expanded [market listings](https://docs.sai.fun/resources/blogs/intro-to-sai#new-products-and-yield:~:text=Real%20world%20asset%20markets) that include real-world asset-style instruments, and DeFi integrations connecting SLP vaults to swaps and routing protocols. The [account](https://docs.sai.fun/resources/blogs/intro-to-sai#new-products-and-yield:~:text=with%20one%20click-,Accounts%3A,-A%20true%20CEX) experience is also evolving. Gasless trades are already live, and the team plans to add multichain and fiat funding routes, [cross-chain USDC deposits](https://docs.sai.fun/resources/blogs/intro-to-sai#new-products-and-yield:~:text=Cross%20chain%20and%20fiat%20funding) via improved on-ramps, and a [mobile application](https://docs.sai.fun/resources/blogs/intro-to-sai#new-products-and-yield:~:text=or%20bank%20account-,Mobile%20app%3A,-Release%20a%20mobile).\n\nFor developers, Sai plans to ship a [data platform](https://docs.sai.fun/resources/blogs/intro-to-sai#new-products-and-yield:~:text=the%20Sai%20platform-,Data%20platform,-%3A%20Backtesting%2C%20historical%20data) with historical data access, backtesting tools, and support for custom order types.\n\n## Closing Summary\n\nSai's Let's Go Saicho trading competition reached 134 traders, generated $6 million in volume, grew TVL 14x in under a month, and collected $21,560 in fees. The two-phase incentive structure produced distinct behavioral shifts, with liquidation rates halving and position sizes shrinking as traders adapted from ROI-focused to volume-focused trading. A 58.2% retention rate in Phase 2 suggests the platform established a sustained baseline of recurring users, though new trader acquisition dropped sharply after the Phase 2 launch spike.\n\nThe next phase for Sai begins after the competition ends. Whether the platform can retain its active traders without incentives, grow SLP liquidity organically, and begin delivering on a roadmap spanning Perps-as-a-Service, Sai Savings, real-world asset markets, and a mobile application will determine whether Sai converts early traction into lasting positioning within the perp DEX landscape.",
"hook": "Sai launched on Nibiru with a two-phase trading competition that attracted 134 traders and generated $6 million in perpetual volume in under a month.",
"publishDate": "2026-04-06T13:50:00Z",
"readingTimeInMinutes": 10.556666666666667,
"slug": "sai-bootstrapping-a-perp-dex-on-nibiru",
"subscriptionTier": "free",
"summary": "Sai is a perpetual DEX on Nibiru offering up to 100x leverage with oracle-settled pricing and pooled SLP-backed liquidity. Its public launch coincided with Let's Go Saicho, a $25,500 competition split between a PnL phase and a volume phase, each targeting different trader behavior. TVL grew 14x to $47,210, cumulative fees hit $21,560, and returning traders accounted for 58.2% of Phase 2 daily activity. The effective fee rate diverged sharply between phases, from 0.64% in Phase 1 to 0.07% in Phase 2, reflecting a shift from high-conviction directional bets to smaller, higher-leverage volume plays. Sai's roadmap extends into Perps-as-a-Service, yield products, RWA markets, and a mobile app, positioning it as trading infrastructure rather than an isolated venue.",
"tags": [
{
"id": "8f138ee0-fb4f-42b8-919c-9feffd558d88",
"name": "Perp DEX"
},
{
"id": "3fdf6b6b-cfac-426d-b20f-3e4794dec982",
"name": "Pulse Reports"
}
],
"title": "Sai: Bootstrapping a Perp DEX on Nibiru",
"previewImage": "https://cdn.sanity.io/images/2bt0j8lu/production/a4aec153be082d301a6690dc37248c9c4ed649d5-6068x3560.png"
},
{
"id": "283895eb-251b-416c-99b1-ed5c3cb67f9c",
"createdAt": "2026-03-29T21:09:41Z",
"updatedAt": "2026-03-31T16:11:30Z",
"assetIds": [
"3748291f-99cb-4002-9bf2-1a23edc0af77"
],
"assets": [
{
"id": "3748291f-99cb-4002-9bf2-1a23edc0af77",
"name": "Nexus Labs",
"symbol": "NEX",
"slug": "nexus-labs"
}
],
"authors": [
{
"id": "25fb4f17-1b40-4477-8e6b-be982a87b349",
"name": "Alice Hou",
"image": "https://cdn.sanity.io/images/2bt0j8lu/production/57dad35450be1a47f2ff6628e6f5fca213499ab4-400x400.jpg?w=100",
"linkedinUrl": ""
},
{
"id": "ac138b81-4baf-4a63-8dbc-e8592c93ba17",
"name": "Matt Kreiser",
"image": "https://cdn.sanity.io/images/2bt0j8lu/production/a09ff5484ce95234175e5e49131ab71bd1b1726d-1024x1024.png?w=100",
"linkedinUrl": ""
}
],
"content": "## Key Insights\n\n* Nexus embeds **high-performance financial engines directly into the protocol through its co-processor model**, moving exchange, margin, and liquidation logic from contract-level simulation into native execution.\n* The **dual-execution architecture allows performance-critical financial workloads and programmable smart contracts to operate in parallel**, avoiding the typical tradeoff between latency and composability.\n* **The Nexus zkVM anchors execution to cryptographic proofs** rather than full validator re-execution, positioning **proof verification as the primary mechanism for scalable correctness.**\n* **With mainnet and exchange deployment in 2026**, Nexus shifts from infrastructure buildout to market validation, where **liquidity formation, and proof-generation efficiency will determine long-term viability. Central to achieving market validation are USDX, the native U.S. dollar stablecoin of the Nexus ecosystem, and the Nexus Exchange, a non-custodial, central limit order book (CLOB) embedded directly into the Nexus Layer 1 (L1).**\n* **The Nexus Exchange is designed to deliver CEX-parity performance in a more self-custodial, verifiable environment that can host and leverage high-frequency trading strategies, AI agents, commerce, and other economic activity.**\n\n## Introduction\n\nFinancial infrastructure depends on complex computational systems that remain largely unverifiable to external participants. Risk models, margin calculations, settlement logic, and internal reconciliation processes operate behind institutional boundaries, requiring users to rely on reporting and oversight rather than direct proof. While public blockchains introduced deterministic execution and transparent state transitions, most existing architectures are not designed to support the performance and computational demands of modern financial markets.\n\nGeneral-purpose chains prioritize composability and shared liquidity but face constraints around latency and throughput. [Application-specific chains](https://messari.io/copilot/share/application-specific-chains-d7e77c8d-2068-4560-84a1-c23e800490dc) achieve higher performance by narrowing scope, yet fragment liquidity and isolate execution environments. In both cases, critical financial logic often remains either offchain or insufficiently optimized for high-frequency, computation-heavy workloads.\n\n[Nexus](https://messari.io/project/nexus-labs) introduces a layered architecture that separates execution, verification, and consensus into independently optimized systems. Its [dual execution model](https://docs.nexus.xyz/architecture/dual-block-execution) combines an [EVM-compatible environment](https://messari.io/copilot/share/evm-compatibility-df970104-0577-49f5-a15e-647f9e746f32) with a specialized financial [co-processor](https://messari.io/copilot/share/co-processor-definition-6dd4fd0e-6438-4882-8403-5fa2234b1ae2), while a native [zkVM](https://messari.io/copilot/share/zkvm-explained-81e8af21-442d-44b7-9201-1a191baf51ef) generates proofs of correct execution that are committed to the base layer. This structure is designed to support performance-sensitive financial applications without relying on external verification frameworks. For example, the [upcoming](https://blog.nexus.xyz/introducing-the-nexus-dex-alpha/) [Nexus ](https://app.nexus.xyz/trade)Exchange is designed to deliver CEX-parity performance in a more decentralized environment that can host and leverage high-frequency trading strategies, AI agents, commerce, and other economic activity.\n\nThis Initiation of Coverage (IOC) report focuses on a technical examination of Nexus’s architecture. Readers seeking a broader discussion of the long-term vision behind verifiable finance are encouraged to refer to the [Nexus Pulse Report](https://messari.io/report/nexus-a-framework-for-verifiable-finance).\n\n[Website](https://nexus.xyz/) / [X (Twitter)](https://x.com/NexusLabs) / [Discord](https://discord.com/invite/AmRKShJfq6)\n\n## Background\n\n[Nexus](https://messari.io/project/nexus-labs/fundraising) was founded in 2022 by [Daniel Marin](https://www.linkedin.com/in/danielmarinq/), a computer science graduate from Stanford University, with the objective of building a universal verifiable machine capable of proving arbitrary computation. The project initially focused on advancing zero-knowledge (zk) proving infrastructure before refining its scope toward financial applications that demand both high performance and computational integrity.\n\nIn June 2024, Nexus [raised](https://blog.nexus.xyz/series-a/) a $25 million Series A round co-led by [Lightspeed](https://messari.io/organization/lightspeed-venture-partners) and [Pantera](https://messari.io/organization/pantera-capital), bringing total capital raised to over $27 million. The network [launched](https://blog.nexus.xyz/the-new-nexus-testnet-is-live/) its first public testnet in December 2024 and has since iterated toward a production-ready architecture.\n\n## Technology\n\n\n\nNexus is structured as a [three-layer architecture](https://docs.nexus.xyz/) composed of an Execution Layer, a Verification Layer, and a Consensus Layer. Each layer operates as a distinct system, responsible for running application logic, generating and validating execution proofs, and finalizing state transitions.\n\n### Execution Layer: Nexus EVM and NexusCore\n\n#### Dual-Block Execution\n\nNexus implements a [dual-block execution model](https://docs.nexus.xyz/architecture/dual-block-execution) designed to separate high-frequency financial processing from general-purpose smart contract coordination. Instead of batching all activity into a single block cadence, the network operates two synchronized block streams with distinct performance characteristics.\n\n[NexusCore](https://docs.nexus.xyz/architecture/nexuscore) targets block times of five milliseconds, while NexusEVM has a block time of 1-2 seconds. These blocks are optimized for latency-sensitive workloads such as order matching, position updates, and risk recalculations. [NexusEVM](https://docs.nexus.xyz/architecture/nexusevm) operates on a slower block cadence, aggregating state changes from several NexusCore blocks before finalization. This periodic synchronization layer preserves compatibility with Ethereum-style smart contracts while allowing composability between programmable logic and high-speed financial activity.\n\nThis structure has three key effects. First, high-speed trading activity runs independently from complex smart contract logic, so time-sensitive operations are not slowed down by heavier computation. Second, performance can scale more efficiently with hardware, as financial workloads do not need to wait for full EVM block processing. Third, both execution environments remain economically unified, with fees and incentives settled at the base layer rather than split across separate systems.\n\nNexusCore\n\nAt the center of NexusCore is the co-processor model. A co-processor is a native execution module embedded into the blockchain itself. Instead of interpreting [smart contract bytecode](https://blog.chain.link/what-are-abi-and-bytecode-in-solidity/), it runs pre-compiled logic with direct access to protocol resources. This design reduces execution overhead and allows financial operations to be processed with greater consistency and speed.\n\nEach co-processor functions as an independent state machine. It maintains its own isolated data structures, executes specialized algorithms tailored to its purpose, and exposes controlled interfaces for interaction. This isolation enables parallel execution across modules while maintaining deterministic state transitions under shared consensus validation.\n\n\n\nThe architecture can be understood in three components:\n\n* **State Layer:** Maintains dedicated data structures and deterministic updates for each co-processor.\n* **Machine Layer:** Executes specialized financial logic, such as matching, margining, or settlement.\n* **I/O Layer:** Enables co-processors to be accessed directly by offchain systems or by smart contracts onchain. \n\nNexusCore’s dual interface is a key advantage. It allows professional trading systems to connect directly for speed, while smart contracts can interact with the same engine onchain. In practice, both high-frequency trading and DeFi applications can run on the same network without sacrificing performance or composability.\n\nOver time, NexusCore is intended to host a broader catalog of “L1-native engines,” including lending markets, vault strategies, oracle and information feeds, RWA and stablecoin infrastructure, gas and fee modules, and bridging primitives, all of which compose atomically with NexusEVM smart contracts.\n\n#### NexusEVM\n\nAt the protocol level, NexusEVM adheres to the standard [Ethereum Virtual Machine ](https://ethereum.org/developers/docs/evm/)specification. It supports the same contract bytecode, gas semantics, RPC interfaces, and developer tooling used across Ethereum.\n\n\n\nWithin Nexus, NexusEVM runs in parallel with NexusCore. Smart contracts deployed on NexusEVM can invoke Core-level co-processors through EVM precompiles or structured cross-domain calls. These interactions are [atomic](https://www.investopedia.com/terms/a/atomic-swaps.asp), meaning that if any part of the transaction fails, the entire operation reverts. Ordered processing ensures deterministic state transitions across validators.\n\nThis integration allows developers to combine programmable contract logic with high-performance financial engines. Applications can manage governance, token logic, incentives, or strategy layers in NexusEVM, while delegating performance-critical execution to NexusCore. In practice, NexusEVM provides an expressive, composable layer of the system, enabling developers to extend and build on top of Core-level financial primitives without leaving the base chain.\n\n### Verification Layer: Nexus zkVM\n\nThe Verification Layer is powered by the Nexus zkVM, a zk virtual machine that generates proofs confirming that computation was executed exactly as specified. Instead of every validator replaying complex logic, the network verifies a succinct proof derived from that execution.\n\nThe zkVM is composed of four primary technical layers:\n\n* RISC-V Machine Architecture: A custom-built virtual machine implementing a modified [RISC-V instruction set](https://messari.io/copilot/share/risc-v-b8a3d4a4-438c-487a-a4ad-53fde178e4f0). It is designed specifically for prover efficiency, including structured memory handling and a “prove only what is accessed” model that reduces unnecessary proof overhead.\n* Algebraic Constraint System (AIR): The execution of the machine is translated into a mathematical representation known as an [Algebraic Intermediate Representation](https://docs.nexus.xyz/zkvm/overview/architecture#core-components). This formalizes every instruction, memory read, and state transition into constraints that must be satisfied for a proof to be valid.\n* STARK-Based Prover (S-two Integration): Execution traces are converted into cryptographic proofs using a [STARK prover](https://github.com/starkware-libs/stwo) optimized for performance. STARKs allow proofs to remain succinct and publicly verifiable without trusted setup requirements.\n* Runtime & SDK Layer: A [Rust](https://rust-lang.org/)-based runtime that allows developers to define public inputs, private inputs, outputs, and logging in a structured way, while abstracting the underlying proving complexity.\n\nNexus zkVM’s benefit is architectural scalability: computation can scale independently from consensus because validators verify proofs rather than replay entire workloads. This reduces replication cost while maintaining deterministic correctness.\n\nA potential risk, on the other hand, is proving overhead. Generating STARK proofs is computationally intensive and requires specialized hardware or distributed prover infrastructure. While verification is lightweight, the economic viability of large-scale proving depends on continued optimization and network-level prover coordination.\n\n### Consensus Layer: Nexus BFT\n\nThe Consensus Layer is governed by [NexusBFT](https://docs.nexus.xyz/), the protocol responsible for finalizing blocks, validating execution commitments, and managing the lifecycle of co-processors.\n\nEach block finalized by NexusBFT includes three core elements:\n\n* A [Merkle commitment](https://messari.io/copilot/share/merkel-commitment-d677ea83-4c90-41b6-b851-bd75041acea5) to the execution state, anchoring the verified results of both NexusCore and NexusEVM.\n* Validator signatures and metadata, establishing agreement across the network.\n* Optional registry updates, which modify the active set of co-processors through the [CPRegistry](https://docs.nexus.xyz/network/overview/system-overview#consensus-layer).\n\nBeyond standard block finalization, NexusBFT introduces protocol-level extensibility. Rather than requiring [hard forks](https://messari.io/copilot/share/what-are-hard-forks-5a6fa661-fb57-4817-9208-aac62b515239) to introduce or modify financial engines, co-processor registration and lifecycle management are handled directly within the consensus layer. This allows the network to activate, upgrade, or deprecate specialized modules without disrupting execution environments.\n\n## DeFi on Nexus\n\n### Strategy and USDX\n\nThe Nexus team is building a native stablecoin and perpetuals-focused Exchange directly into the Layer 1, given that these are [three of the most](https://blog.nexus.xyz/usdx-the-money-layer-for-verifiable-finance/#:~:text=Ethereum%2C%20and%20beyond.-,The%20USDX%20thesis,-At%20the%20core) extensible, proven and synergistic businesses in crypto. Taken together, the L1 and exchange are designed to deliver CEX-parity performance in a more decentralized environment that can host and leverage high-frequency trading strategies, AI agents, commerce, and other economic activity.\n\nUSDX is the [native U.S. dollar stablecoin](https://blog.nexus.xyz/usdx-the-money-layer-for-verifiable-finance/) of the Nexus ecosystem, 1:1-backed by U.S. Treasury bills and cash equivalents. USDX will be the default margin and settlement asset for [Nexus ](https://app.nexus.xyz/trade)Exchange, a non-custodial [central limit order book](https://messari.io/copilot/share/central-limit-order-book-0364d6bc-f9e7-4059-99e7-28066b7cc1f5) (CLOB) embedded directly into the Layer 1 (L1). The more the Exchange is used, the greater the demand for USDX. Moreover, the more USDX in circulation, the deeper and cheaper liquidity becomes for the exchange. With meaningful adoption, this reinforcing flywheel can increase the value of the L1.\n\nUSDX will first launch on Ethereum, where distribution is deepest, followed by Nexus mainnet, and crosschain interoperability. Issuance will take place via the M0 Protocol via the JMI extension, which will enable permissionless 1:1 swaps from major stablecoins into USDX. As tokenized assets and 24/7 synthetic markets expand onchain, USDX is designed to serve as the neutral denominator for risk-managed portfolios.\n\nThe key value proposition of USDX relative to other stablecoins is [yield streaming](https://blog.nexus.xyz/yield-is-the-2026-defi-battleground/#:~:text=DeFi%20in%202026.-,Yield%20streaming%3A%20from%20extraction%20to%20alignment,-Yield%20streaming%20inverts), which inverts the [extractive model](https://blog.nexus.xyz/yield-is-the-2026-defi-battleground/#:~:text=The%20extractive%20model%3A) of the largest stablecoins USDT and USDC, whose issuers capture all reserve yield as compensation for infrastructure and compliance. Nexus streams USDX yield directly to users and builders on the L1 via the fully transparent Global Yield Distribution System ([GYDS](https://blog.nexus.xyz/usdx-the-money-layer-for-verifiable-finance/#:~:text=Nexus-,Yield%20streaming%20and%20the%20Global%20Yield%20Distribution%20System,-One%20of%20the)). Yield from U.S. Treasury bills and cash equivalents is distributed as USDX each week according to the time-weighted USDX balances across registered application sources, which are contracts and modules that builders have opted into. The yield is split between the protocol and builders, with builder allocations proportional to attributed USDX TVL, and yield flowing through to end users based on their balances, where supported. On top of this, Nexus can, at its discretion, add protocol-native incentives to increase yield beyond short-duration U.S. Treasury yields, including portions of exchange revenue, subject to governance and risk budgets, directed at strategic segments like builders bringing new markets or liquidity. The Nexus team plans to publish a quarter-by-quarter policy for USDX, disclosing a target onchain yield that reflects the baseline Treasury rate plus any protocol-native additions.\n\nThis aligns incentives to a much higher degree than Tether’s USDT and Circle’s USDC, which simply offer access to the stablecoin as the value proposition. It’s also a process entirely unblocked compared to yield-bearing stablecoins registered as securities (which severely limits the holder base to qualified participants) in order to pay interest directly to holders. \n\nThe yield streaming model of USDX creates a predictable, programmatic incentive loop. Builders are rewarded for attracting real USDX usage, while users are rewarded for holding and deploying USDX on Nexus. During the early phase, NEX token incentives may complement USDX yield to accelerate integrations, but the long-term design relies on organic demand and real yield.\n\n### Nexus Exchange\n\nTo operationalize its execution architecture, Nexus [is developing](https://blog.nexus.xyz/introducing-the-nexus-dex-alpha/) the [Nexus ](https://app.nexus.xyz/trade)Exchange, a non-custodial [central limit order book](https://messari.io/copilot/share/central-limit-order-book-0364d6bc-f9e7-4059-99e7-28066b7cc1f5) (CLOB) embedded directly into the Layer 1 (L1). Unlike exchanges deployed as smart contracts, the Nexus Exchange runs inside NexusCore as a native co-processor. Order matching, margin calculations, funding logic, and liquidations are executed at the protocol level rather than simulated through contract bytecode.\n\nThe first supported product is [perpetual futures](https://docs.nexus.xyz/trading/perpetuals). These contracts allow traders to take leveraged long or short exposure to supported assets without expiration. Users post collateral, open positions with leverage, and pay or receive periodic funding based on market conditions. Because the exchange engine operates within NexusCore, all position updates and risk calculations are processed under deterministic execution and shared consensus.\n\nThis architecture shifts the exchange from being an application layered on top of the chain to becoming part of the chain’s execution fabric. Performance-sensitive operations are handled natively, while settlement and accounting remain transparent and verifiable at the base layer.\n\n### Risk Management and Liquidation\n\nRisk controls are enforced automatically through a built-in [liquidation engine](https://docs.nexus.xyz/trading/perpetuals/liquidations). A trader’s equity is continuously evaluated against predefined margin thresholds using a mark price derived from the Nexus oracle system.\n\nTwo margin levels govern positions. Initial margin determines the leverage required to open a position, while maintenance margin defines the minimum equity needed to keep it open. If equity falls below the maintenance threshold, the system triggers liquidation.\n\nRather than relying on external bots competing to liquidate positions, Nexus executes liquidations through a dedicated onchain mechanism. Positions are closed at or near the mark price, with safeguards to prevent negative balances. If losses exceed available collateral, an [insurance fund](https://docs.nexus.xyz/trading/perpetuals/liquidations#how-liquidation-works) absorbs residual shortfalls to preserve overall system solvency.\n\nBy embedding margin and liquidation logic directly into the execution layer, Nexus reduces execution uncertainty and race conditions. The tradeoff is that core risk parameters are embedded at the protocol level, making changes more consequential than in contract-based systems.\n\nTogether, the Nexus Exchange functions as both a flagship application and a structural demonstration of NexusCore’s capabilities. It tests whether protocol-level financial engines can combine high-performance execution with deterministic settlement and cryptographic accountability within a unified L1 environment. Beyond perpetual futures, Nexus plans to expand the Exchange to include [spot markets](https://docs.nexus.xyz/trading/spot) and [vault products](https://docs.nexus.xyz/trading/vaults), with further details expected in future releases.\n\n## Roadmap \n\nNexus’s [roadmap](https://blog.nexus.xyz/the-nexus-roadmap/) outlines a transition from testnet infrastructure to a fully operational financial L1, with staged activation of validators, exchange functionality, and protocol-level financial primitives.\n\n### Q1 2026: Network Activation\n\nThe first milestone in 2026 is under active development and will center on network coordination and exchange readiness. [Community Genesis](https://blog.nexus.xyz/tag/community/) will onboard validators, operators, and early participants, marking the shift from a development-driven network to a validator-secured environment.\n\n### Q2 2026: Mainnet EVM Launch\n\nIn Q2 2026, the Nexus L1 mainnet is expected to go live. This milestone will establish a production settlement layer with secure execution and finalized consensus. Bridges and onramps will become operational, allowing external capital to enter the ecosystem.\n\nThis stage will formalize validator participation and transition the EVM environment from testnet to persistent infrastructure. Applications will be able to deploy into a stable environment with deterministic finality and integrated access to NexusCore co-processors. The focus will shift from experimentation to economic durability.\n\n### Q3 2026: Exchange Mainnet\n\nFollowing L1 activation, the Nexus Exchange will launch on mainnet. This will mark the operational start of Nexus as a functioning financial network rather than solely an infrastructure layer. The exchange is designed to deliver CEX-like performance in a more decentralized environment that can host and leverage high-frequency trading strategies, AI agents, commerce, and other economic activity.\n\nLive trading will introduce continuous order flow, real margin enforcement, oracle updates under production conditions, and full interaction between execution, verification, and consensus layers. Liquidity depth, liquidation behavior, and proof performance will become measurable under sustained market activity.\n\nBeyond launch milestones, Nexus will continue to evolve its core architecture. Development efforts will focus on improving zkVM performance, expanding co-processor capabilitiesand broadening supported asset classes such as 24/7 equities, FX, commodities, and indexes.\n\nThe long-term objective will be to extend protocol-level financial infrastructure across additional markets, collateral models, and composable applications. This phase will prioritize improvements in proof efficiency, execution reliability, and validator robustness to support sustained financial activity.\n\n## Closing Summary\n\nNexus represents an architectural bet: that high-performance financial infrastructure should not be simulated through smart contracts, but embedded directly into the base layer and verified cryptographically.\n\nIts three-layer design separates execution, proof generation, and consensus, allowing each to scale independently. The dual-execution model formalizes a distinction between programmable logic and performance-critical financial engines. NexusCore handles deterministic, latency-sensitive computation, while NexusEVM preserves composability and developer accessibility. The zkVM anchors the system with verifiable computation, shifting validation from re-execution to proof verification.\n\nWith mainnet and exchange deployment in 2026, Nexus shifts from infrastructure buildout to market validation, where liquidity formation, oracle reliability, and proof-generation efficiency will determine long-term viability. Central to achieving market validation are USDX, the native U.S. dollar stablecoin of the Nexus ecosystem, and the Nexus Exchange, a non-custodial, central limit order book (CLOB) embedded directly into Layer 1 (L1). USDX inverts the extractive model of the largest stablecoins USDT and USDC, whose issuers capture all reserve yield as profit for infrastructure and compliance. Nexus streams USDX yield directly to users and builders on the L1 via the fully transparent Global Yield Distribution System (GYDS). The Nexus Exchange is designed to deliver CEX-parity performance in a more decentralized environment that can host and leverage high-frequency trading strategies, AI agents, commerce, and other economic activity.\n\nBy integrating matching, margining, liquidation, and oracle logic at the protocol level, Nexus reduces overhead and execution uncertainty inherent in contract-based exchange designs. The tradeoff is structural: financial logic becomes part of the base layer, increasing the importance of validator coordination and disciplined protocol governance.\n\nUltimately, the success of Nexus will not be measured by throughput alone, but by whether protocol-level financial primitives can operate reliably under live market conditions while maintaining deterministic settlement and cryptographic accountability.",
"hook": "Nexus embeds high-performance financial engines directly into the protocol through its co-processor model, moving exchange, margin, and liquidation logic from contract-level simulation into native execution.",
"publishDate": "2026-03-31T14:00:00Z",
"readingTimeInMinutes": 14.376666666666667,
"slug": "understanding-nexus-a-comprehensive-overview",
"subscriptionTier": "free",
"summary": "The dual-execution architecture allows performance-critical financial workloads and programmable smart contracts to operate in parallel, avoiding the typical tradeoff between latency and composability. The Nexus zkVM anchors execution to cryptographic proofs rather than full validator re-execution, positioning proof verification as the primary mechanism for scalable correctness. With mainnet and exchange deployment in 2026, Nexus shifts from infrastructure buildout to market validation, where liquidity formation, and proof-generation efficiency will determine long-term viability. Central to achieving market validation are USDX, the native U.S. dollar stablecoin of the Nexus ecosystem, and the Nexus Exchange, a non-custodial, central limit order book (CLOB) embedded directly into the Nexus Layer 1 (L1).",
"tags": [
{
"id": "2df3c948-022e-4a95-9ecf-d2ddca4c40f3",
"name": "Protocol Overview"
}
],
"title": "Understanding Nexus: A Comprehensive Overview",
"previewImage": "https://cdn.sanity.io/images/2bt0j8lu/production/6c006611ca5148c732fac20be7f0f8ecc6641368-6400x3600.png"
}
],
"metadata": {
"limit": 10,
"page": 1,
"totalRows": 2,
"totalPages": 1
}
}{
"data": "<unknown>",
"error": "<string>"
}{
"data": "<unknown>",
"error": "<string>"
}{
"data": "<unknown>",
"error": "<string>"
}{
"data": "<unknown>",
"error": "<string>"
}Authorizations
Query Parameters
Limit must be less than or equal to 1000
Example: 4515ba15-2719-4183-b0ca-b9255d55b67e
Example: stablecoins, defi
Whether to include full report content in the response
⌘I