Cross-chain Swaps on TON: Symbiosis, Allbridge, and LayerZero

A cross-chain swap is the movement of value between blockchains without forcing the user to chain separate “bridge → swap → bridge” operations — a single UX instead. On TON as of May 2026 three main protocols deliver that experience: Symbiosis, Allbridge (in two forms — Core and Classic), and LayerZero (via the OFT standard). This article unpacks how they work, where trust models diverge, what fees and risks look like — and when each tool makes sense.

Why cross-chain matters on TON

TON is not an EVM chain, and there is no native “Ethereum-to-Arbitrum”-grade bridge. To bring stablecoins, ETH, BTC, and other assets into TON DeFi, a bridge is required.

Three base tasks:

1. Move USDC/USDT from L1/L2 into a TON form for DEX and lending protocols.
2. Get ETH or WBTC in a form usable in TON DeFi.
3. Swap an asset on one chain directly into an asset on another (e.g., ETH → TON) without manual steps.

Each protocol solves these differently.

Symbiosis: a cross-chain liquidity protocol

Symbiosis targets universal cross-chain swaps across dozens of chains, including TON.

Mechanics:

• Relay-node network and liquidity pools in each connected chain.
• User signs a transaction on the source chain; funds enter the pool; on the TON side the equivalent is paid out via a local pool.
• Final swap into the desired jetton runs through integrated DEXes (STON.fi, DeDust).

Trust model: Symbiosis validator multi-sig + liquidity pools. Not a canonical bridge — closer to a networked exchanger with its own liquidity.

Strengths: one UI for dozens of pairs, fairly fast (minutes), integrated into popular aggregators.

Weaknesses: depends on per-chain pool depth — for rare pairs slippage is noticeable. Trust model is multi-sig, not L1-secured.

Allbridge: Core vs Classic

Allbridge is among the oldest bridges in the TON ecosystem and ships in two forms.

Allbridge Core (message-based)

Modern architecture — cross-chain messages via its own validator set + integration with other messaging protocols. Supports atomic cross-chain swaps (not just asset transfer, but simultaneous conversion). Trust model is close to Symbiosis — multi-sig.

Allbridge Classic (canonical bridge model)

Traditional canonical bridge: on the source chain the asset is locked in a contract, on TON a wrapped version is minted as a jetton. On the return path — burn and unlock.

Trust model here is a validator set confirming lock/unlock in both directions. Effectively a federation rather than a trust-minimized bridge.

Strengths: long-standing TON presence, proven infrastructure, two modes for different needs.

Weaknesses: canonical wrappers historically have proven vulnerable across the industry (Multichain 2023, Wormhole 2022 — category-level, not Allbridge specifically); TON in 2026 already saw the TAC drain — worth keeping in mind.

LayerZero and the OFT standard

LayerZero is an omnichain messaging protocol serving hundreds of projects. For tokens it introduced the OFT (Omnichain Fungible Token) standard:

• The same token “lives” across multiple chains but has a single supply via mint/burn.
• Moving from chain A to chain B is a burn on A + mint on B with cryptographic confirmation over the LayerZero channel.
• On TON, OFT tokens are implemented as jettons tied to a LayerZero channel.

Trust model: oracle + relayer set chosen by the OFT issuer. That gives flexibility — but security depends on the issuer’s choice.

Usage on TON as of May 2026: Ethena’s USDe, plus a series of projects with multi-chain deployment. LayerZero v2 runs stably, but it’s infrastructure — the profile of each specific OFT issuance is critical.

Comparison table

Fees and slippage

Cross-chain swaps cost more than same-chain swaps — that is expected. Cost breakdown:

1. Source-chain gas. Ethereum at peak: $5–30. L2s: cents.
2. TON gas. Tens of cents for final operations (mint jetton, swap).
3. Bridge fee. 0.05–0.3% of amount depending on protocol.
4. Swap slippage. For large amounts in thin pairs 0.5–3%.
5. Spread between wrapped and native forms. Sometimes bridged-USDC trades at a 0.1–0.5% discount vs. native USDT — a tax on wrapper liquidity.

For $100 the cost-of-bridge can be $1–3 from Ethereum, for $10,000 — $20–60, so small amounts are disproportionately expensive in relative terms.

Risks: what actually happened in the industry

Bridges have historically been the most vulnerable infrastructure class in DeFi. Key incidents — for scale, not as direct critique of the three protocols here:

• Nomad Bridge (August 2022). An initialization mishap allowed $190M to be drained. Lesson: formal verification of initializations.
• Wormhole (February 2022). A signature exploit minted 120k wETH from nothing — $326M. Lesson: validator signature checking.
• Multichain (July 2023). Founder key compromise — $130M+ left bridges. Lesson: key custody centralization.
• TAC Bridge on TON (2026). $2.5M via exploit — the freshest case relevant to the TON ecosystem.

!

Main rule. Do not park large amounts in bridged form. If you hold $10k+ on TON in bridged USDC, prefer native USDT (Tether’s TON issuance, not bridged) or split between two independent bridges.

In practice: ETH → TON via three paths

Scenario: move 1 ETH from Ethereum mainnet to TON and buy TON jettons.

Path A — Symbiosis

1. Open the Symbiosis interface, select ETH → TON.
2. Sign on Ethereum (gas $5–15 in a calm hour).
3. In 1–3 minutes receive native TON equivalent.
4. Swap to the desired jetton in the same UI or on STON.fi.

Pros: single UI. Cons: depends on Symbiosis liquidity.

Path B — Allbridge

1. On Allbridge select ETH from Ethereum → wETH on TON.
2. Receive the wETH jetton.
3. On STON.fi/DeDust swap wETH to TON or USDT.

Pros: proven infra. Cons: ETH stays wrapped until the final swap.

Path C — LayerZero OFT (if ETH is available as OFT)

1. Through an app supporting an ETH OFT (as of May 2026 a native “ETH OFT” does not exist; typically stETH/wstETH variants for liquid staking are used).
2. Burn on Ethereum, mint OFT jetton on TON.
3. Swap on a DEX.

Pros: unified supply, no wrapper as a separate entity. Cons: not all assets are available as OFT.

Decision matrix

When to pick which:

• Small amounts, rare assets: Symbiosis (versatility + UX).
• USDC/USDT, mid-sized amounts: Allbridge Core (atomic swaps).
• Long-term TON holding: move into native form (native USDT on TON, native TON) — don’t sit in bridged.
• OFT-supporting tokens (Ethena USDe, etc.): LayerZero — because the supply is unified, not a wrapper.
• Large amount (>$50k): split between two independent bridges + move into native form.

What this means for users: a checklist

1. Don’t bridge a large amount on the first try — test with $10–50 first.
2. Verify you receive a native jetton, not a wrapper, if you plan to hold.
3. Compare quotes across Symbiosis, Allbridge, and aggregators before swapping — spreads can be noticeable.
4. Watch bridge status — Twitter/Telegram are the fastest incident channels.
5. If a bridged asset trades at a 1%+ discount to native — that is a liquidity/trust signal.
6. Large amounts deserve recipient-contract verification before the operation.

Where the industry is heading

Several directions for 2026–2027:

• Native USDC on TON. Circle has repeatedly hinted at non-EVM expansion. Native issuance would erode demand for bridged wrappers.
• OFT standardization. LayerZero v2 and peers aim to make omnichain the default — that reduces demand for classic canonical bridges.
• Bridge audits and insurance. Services (Nexus Mutual, analogs) now insure specific bridges — changing risk/reward for whales.
• Regulatory focus. EU MiCA already indirectly regulates cross-chain stablecoin issuance through issuers; Russia and the EAEU remain open questions.

Conclusion

Cross-chain swaps on TON are working infrastructure — but they reward attention to detail. Symbiosis is convenient for rare pairs and universal UX, Allbridge for USDC/USDT and proven flows, LayerZero OFT for assets designed as omnichain. Baseline discipline: don’t park large amounts in bridged wrappers, split large transfers between independent paths, and verify that the final jetton is the one you intended.

Bridges are the highest-risk infrastructure class in DeFi, and industry history confirms that. Caution costs less here than the post-hoc lesson.

→