Codex is the Ethereum Layer 2 launched in 2024 by the Dragonfly-backed team now branded Codex FX, built to make USDC, USDT, and other dollar tokens first-class assets rather than bolted-on ERC-20s. As of September 2026, the company's commercial focus has shifted almost entirely to off-chain FX and bridging products (Codex FX), and the Codex L2 itself shows $0 in tracked DeFi activity on DeFiLlama. This guide explains what the Codex chain still is on paper, why the company's own site no longer markets it as the flagship product, where it fits among the stablecoin chains that did gain traction (Plasma, Tron), and what that means for builders evaluating it in 2026. Note also that docs.codex.io is an unrelated company (a DeFi data/indexing API), not this Codex, a name collision worth knowing about if you search for developer docs.
​

By the end you will have an accurate, current-as-of-September-2026 comparison across Codex, Plasma, Stable, and Tron, a practical view of when (if ever) the Codex chain is the right choice today, and a sense of how cross-chain orchestration sits on top of any single-chain decision.
​

Codex is an Ethereum Layer 2 rollup launched in 2024 by a team focused exclusively on stablecoin payments infrastructure. The chain uses an EVM-compatible execution layer, posts data and proofs to Ethereum L1, and offers institutions a settlement environment optimized around three properties: predictable fees denominated in stablecoins, native compliance hooks for regulated issuers, and direct integration with mint and redemption rails from issuers like Circle.
​

The pitch is simple. General-purpose chains charge gas in their native token (ETH, SOL, TRX), expose users to volatile fee markets, and treat stablecoins as just another ERC-20. Codex flips that model, stablecoins are the native unit of account, gas can be paid in USDC, and the chain's roadmap is shaped by what payment processors and FX desks actually need rather than what DeFi power users want.
​

The team has been public about defending Ethereum's role as the settlement layer for stablecoin economies, framing Codex as a way to keep regulated dollar volume inside the Ethereum security perimeter rather than ceding it to alternative L1s. The Defiant covered this positioning in detail when the team announced its strategic direction, calling Codex's launch a response to the "stablechain attack" from non-Ethereum stablecoin chains. Coverage from The Defiant on Codex's Ethereum-defense thesis lays out the strategic rationale. For comparison context with the largest dollar-token issuers Codex serves, see the USDC versus Tether comparison.
​

Codex runs as an optimistic rollup with EVM equivalence, meaning Solidity contracts deploy without modification. Transactions are sequenced and executed on Codex, batched, and posted to Ethereum L1 with fraud-proof windows that finalize after the standard challenge period. For most payment use cases, the user experience is faster than vanilla Ethereum L2s because the team optimized the sequencer for predictable throughput on stablecoin transfer patterns rather than long-tail smart contract calls.
​

The architecture has four notable design choices:

The full developer documentation is published on the project's site, and the Ethereum Foundation's Layer 2 overview is the best primer for readers new to rollup architecture.
​

The framing above describes what the Codex chain was designed to do. It is no longer an accurate picture of where the company puts its energy. As of September 2026, codex.xyz markets the business entirely as "Codex FX," a principal-desk stablecoin bridging and cross-border FX service: on/off-ramp, blockchain bridge, and a 1:1 USDC/USDT swap API. The Codex L2 chain itself is not in the site's product navigation at all; it appears only as one of nine settlement networks (alongside Ethereum, Tron, Solana, Polygon, Base, Optimism, Arbitrum, and Avalanche) that the bridging desk can pay out on, per Codex's own bridge page.
​

The on-chain numbers confirm the shift. DeFiLlama's chain-by-chain TVL data lists Codex at $0, the same table showing Plasma, Stable, and a dozen smaller chains all with real, non-zero DeFi activity. That does not mean the Codex chain was deprecated outright, industry profiles like Gate Learn's overview and Crypto-Fundraising's project page still describe Codex Chain, Codex API, and Codex FX as three parts of one stablecoin ecosystem, and the chain still appears as a live settlement option in Codex's own bridge tool. But commercially, Codex FX (the off-chain bridging and FX desk) is what the company leads with, sells, and reports $1B+ in monthly cleared volume against, not the L2.
​

For builders, the practical read is: treat "Codex" today as primarily an FX/bridging counterparty you integrate with via API or demo request, not a chain you deploy contracts to or route production DeFi traffic through. The architecture and design choices described below are what the chain was built to do; whether that chain sees renewed on-chain activity going forward is an open question the company's own current marketing does not answer.
​

The four most-discussed stablecoin chains in 2026 each made different bets. Codex bet on Ethereum security and institutional credibility. Plasma bet on Bitcoin-anchored security and zero-fee retail UX. Stable bet on USDT-as-gas and Tether co-design. Tron is the incumbent, the chain that already moves more retail USDT than anyone else and shows no sign of stopping.
​

Here is the practical comparison matrix.

The differences look stark on the matrix, but the real decision driver is volume distribution. Bitget's analysis of stablecoin chain wars discussed Ethereum's stablecoin dominance over Tron; as of September 2026, Ethereum carries $146.7B in stablecoin market cap to Tron's $94.3B (DeFiLlama), and Codex, by inheriting Ethereum security, nominally sits inside that larger pool, though its own share of it is currently negligible. Bankless's interview with Codex cofounder Haonan Li on onchain FX is the deepest public source on the team's strategic reasoning.
​

Codex's early traction has come from three categories of users. Each one maps to a Codex design choice.
​

FX desks moving USD-to-EUR or USD-to-other-fiat positions across counterparties have used Codex's USDC and EURC support as a settlement venue. The combination of predictable fees, fast L2 confirmations, and Ethereum L1 finality fits the risk-management posture of regulated FX operators better than zero-fee chains, where the ops team has to explain why their settlement rail does not charge them.
​

Marketplaces and platforms paying out to vendors in dollars have used Codex as one leg of a multi-chain payout strategy, especially for European counterparties who prefer EURC. The chain integrates cleanly with the broader category of B2B stablecoin payout APIs as a destination chain.
​

Treasury teams holding stablecoin floats across multiple chains have used Codex as a settlement target when EURC or other regulated euro tokens are part of the mix. The native compliance hooks make audit trails simpler to assemble than on chains where every memo field is custom. Operators handling rebalancing across this kind of multi-chain float typically combine Codex with general-purpose stablecoin rebalancing tools that pick the right destination chain per move.
​

The mistake some teams make when evaluating Codex is treating "pick a stablecoin chain" as the question. In production, the question is "how do we move stablecoins across whatever chains our counterparties use." Codex is one node in a graph that also includes Ethereum mainnet, Tron, Solana, Arbitrum, Base, HyperEVM, Plasma, and a long tail of L2s.
​

This is where the Rail / Layer / App distinction matters. CCTP, Hyperlane, and LayerZero are the rails, the underlying transport protocols that move value between chains. Layers like Eco Routes orchestrate across those rails, picking the cheapest or fastest path for each transfer based on cost, speed, and finality. Apps sit on top of the layer and never need to encode the chain-by-chain logic themselves. For a deeper view of the rail-level options available to orchestration layers, see the cross-chain messaging protocols guide.
​

Codex fits this model cleanly. A treasury platform integrating a stablecoin API at the orchestration layer can route to Codex when EURC or institutional EU flow is in scope, to Tron when retail USDT is in scope, and to Plasma or Stable when fee-sensitive USDT flow is the use case, without writing chain-specific code each time.
​

No chain is a free lunch. Codex's tradeoffs are worth being honest about.
​

Centralization concerns at the sequencer. Like most early-stage L2s, Codex runs a single sequencer. This is the standard tradeoff for fast block production, but it concentrates liveness risk. A decentralized sequencer roadmap is on the project's roadmap, and L2BEAT's tracking of L2 sequencer status is the canonical reference for monitoring this across the L2 ecosystem.
​

Liquidity depth versus incumbents. Tron and Ethereum mainnet still hold the deepest USDT and USDC liquidity respectively. Codex chain liquidity is not growing, per DeFiLlama it currently shows $0 in tracked on-chain activity, so it is not a venue you would route DeFi liquidity through today regardless of trade size. For large flows, orchestration layers split the trade across multiple chains, with Codex as one optional leg.
​

Regulatory positioning is still evolving. Codex's compliance posture is designed for a world where issuers (Circle, regulated euro issuers) are first-class participants. If global stablecoin regulation tightens around chain-level controls, Codex is well positioned. If it loosens or fragments, the value of native compliance hooks becomes less clear.
​

The chain count is multiplying. Codex is one of several stablecoin chains launching in 2025-2026. Stablewatch's review of stablecoin chains tracks the broader category. Builders should plan for a multi-chain world rather than betting on any single venue.
​

For developers building cross-chain stablecoin products, the practical pattern is to integrate at the orchestration layer rather than wire to Codex directly. Eco Routes selects between CCTP, LayerZero, Hyperlane, and other rails to move USDC and USDT across chains, with Codex available as a destination once liquidity reaches the threshold for production routing. The user signs an intent ("send X USDC from Arbitrum to recipient on Codex"), and a solver network competes to fulfill it atomically, no manual bridging, no chain-specific glue code.
​

The Routes CLI exposes the same surface for every supported chain, so adding Codex to a production payout pipeline is the same operation as adding any other chain. Eco currently supports stablecoins across Ethereum, Optimism, Base, Arbitrum, HyperEVM, Plasma, Polygon, Ronin, Unichain, Ink, Celo, Solana, Sonic, BSC, and Worldchain.
​

Codex is an Ethereum Layer 2 rollup, not a standalone Layer 1. It posts transaction data and validity information back to Ethereum mainnet, inheriting Ethereum's security guarantees. This is a key differentiator from Plasma, Stable, and Tron, which are independent L1s with their own consensus and validator sets.
​

Technically yes, it has not been shut down or deprecated, but its commercial priority has clearly shifted. As of September 2026, Codex's own website (codex.xyz) markets the business entirely as "Codex FX," an off-chain bridging and FX desk, and does not list the Codex L2 chain as a current product in its navigation. DeFiLlama shows $0 in tracked DeFi activity on the chain. Builders evaluating Codex today are, in practice, evaluating the FX/bridging product, not the L2.
​

Yes. Codex was designed so users and applications can pay transaction fees directly in USDC. This removes the need to hold ETH for gas, which is one of the main onboarding frictions for enterprise teams running production stablecoin flows. Other chains in the comparison handle this differently, Stable uses USDT for gas, Plasma offers zero-fee USDT, and Tron uses its energy model.
​

Codex prioritizes Ethereum security and institutional credibility, with USDC and EURC as native assets and direct issuer integration. Plasma prioritizes Bitcoin-anchored security and zero-fee retail USDT payments, with deep Tether partnership and a self-custody wallet integration. Codex fits regulated FX and B2B settlement; Plasma fits high-volume retail remittances.
​

USDC is the flagship asset on Codex with native Circle integration. EURC is supported for euro-denominated flows. Additional dollar and FX tokens have been added as issuer partnerships expand. The chain is designed to onboard new regulated stablecoins quickly through its compliance-aware contract framework. The Tether USDT guide covers how the dollar-token landscape splits across chains today.
​

Cross-chain transfers from Codex use the same orchestration layers that handle multi-chain stablecoin movement generally. Eco Routes selects between CCTP, LayerZero, Hyperlane, and other rails based on cost, speed, and the destination chain. Direct bridge integrations are also available for Ethereum mainnet round-trips. The solver networks guide explains how the orchestration layer's competitive fill model actually works.
​

Codex chain was designed as a stablecoin-native Ethereum L2 betting that institutional dollar volume wanted Ethereum security plus a payment-optimized execution environment. As of September 2026, the company's own marketing and product lineup have moved on: Codex FX, an off-chain bridging and FX desk, is now the flagship product, and the chain itself shows $0 in tracked DeFi activity on DeFiLlama. For builders, the right question is not "which stablecoin chain wins" but "how does my orchestration layer route across all of them," and today that means treating Codex chain as a dormant option rather than an active one. Codex FX, the company's current product, is a separate integration decision (an API/bridging counterparty, not a chain to deploy to). Cross-chain orchestration platforms like Eco Routes treat Codex the same way they treat every other supported chain, as a destination the application picks, not a ceiling the application has to wire around.
​

For the underlying taxonomy, see Net vs Gross Settlement: RTGS and DNS Explained.