Most "which chain is cheapest?" comparisons stop at the network fee. That number — $0.0004 on Solana, $2 on Tron without staked energy, $1.50 on Ethereum on a quiet afternoon — is real, but it is not the bill.
The true cost of a stablecoin transfer includes: the network fee itself, the cost to acquire whatever gas token the chain requires, any markup from the wallet or payment service provider processing the transfer, and the off-ramp fee when the stablecoin converts back to local currency. Layer those together and the all-in picture by chain looks substantially different — and the right chain for a $100 remittance is not necessarily the right chain for a $100,000 treasury settlement.
This piece works through the full cost stack and builds three scenarios: a $100 personal transfer, a $1,000 business payment, and a $100,000 treasury or B2B settlement.
The cost stack
Every stablecoin transfer carries some combination of four cost types:
1. Network fee (gas). The fee paid to the blockchain for processing the transaction. This is the number most quoted in chain comparisons. It is denominated in the chain's gas token, which may or may not be the stablecoin you're transferring.
2. Gas-token acquisition cost. On chains where gas is paid in a volatile native token (ETH, SOL, TRX), users who don't already hold that token must buy it. An exchange or on-ramp charges a spread — typically 0.5–2% for a retail conversion, lower for high-volume users. This cost is often invisible but real.
3. PSP or wallet markup. Payment service providers, wallets, and neobanks that abstract gas management for users typically charge a spread on the transfer. This ranges from zero (self-custody with your own gas) to 0.5–2% for managed services.
4. Off-ramp fee. When the recipient converts stablecoins back to local currency — the last step in most cross-border payment flows — the off-ramp provider charges a fee. This varies by corridor, provider, and chain. Ranges: 0.5–3% in competitive corridors; 3–7%+ in underserved markets.
Not every transfer incurs all four costs. A self-custodied wallet-to-wallet transfer between two parties who both hold the same stablecoin and never off-ramp pays only the network fee. A consumer remittance from a US worker to family in West Africa pays all four.
Network fees by chain (mid-2026 benchmarks)
| Chain | Typical network fee (USDC/USDT transfer) | Notes |
|---|---|---|
| Ethereum | $0.50–$5+ | Variable; peaks during congestion |
| Tron | $0–$4 | ~$0 with pre-staked TRX energy; $1.92–$4 without |
| Solana | < $0.001 | Consistently sub-cent; ~$0.0004 typical |
| Base | $0.005–$0.05 | OP Stack calldata cost varies with L1 conditions |
| Tempo | Target < $0.001 | Protocol fee target; $0.001 per Stripe partnership terms |
Network fees are point-in-time estimates subject to network conditions. Ethereum fees sourced from ycharts.com and block explorer medians; Tron from TronNRG and Messari Q1 2026; Solana from Chainspect and Solana docs; Base from docs.base.org; Tempo from official fee target and Stripe partnership blog.
Gas-token friction: the hidden cost
On Ethereum, every transaction requires ETH. A user holding only USDC who wants to send USDC must first acquire ETH — either from an exchange (retail spread: 0.5–1.5%), or from a paymaster service that absorbs the cost and charges it back differently.
At a $100 transfer: if acquiring $2 of ETH costs a 1% exchange spread, that is $0.02 in friction — small, but not zero. More significant is the holding cost: ETH price risk on the gas float. For a business sending 10,000 transfers per month, the ETH float is not trivial and must be managed.
Tron's TRX staking model creates similar friction in a different shape. Staking 5,000 TRX (approximately $1,200 at Q1 2026 prices) removes the per-transfer fee, but locks up capital with price exposure. Businesses with steady volume can model this as a capital cost: opportunity cost of staked TRX plus TRX price variance.
Solana's SOL fees are so small — fractions of a cent — that gas-token acquisition friction is negligible in practice. Most wallets handle this invisibly.
Tempo has no gas token. Gas is paid in the stablecoin being transferred, via the protocol's Fee AMM. There is no ETH to buy, no TRX to stake, no SOL float to maintain. The gas-token acquisition cost is zero by design.
Scenario 1: $100 personal transfer (remittance)
A worker in the United States sends $100 to family in the Philippines, using a self-custody wallet on each end. The recipient off-ramps to local currency.
| Chain | Network fee | Gas-token cost | PSP/wallet | Off-ramp (est. 2% corridor) | Total | % of transfer |
|---|---|---|---|---|---|---|
| Ethereum | $1.50 (mid) | $0.02 | — | $2.00 | ~$3.52 | ~3.5% |
| Tron (no stake) | $3.00 (mid) | — | — | $2.00 | ~$5.00 | ~5.0% |
| Tron (pre-staked) | ~$0 | (capital cost) | — | $2.00 | ~$2.00 | ~2.0% |
| Solana | $0.001 | ~$0 | — | $2.00 | ~$2.00 | ~2.0% |
| Base | $0.02 (mid) | $0.01 | — | $2.00 | ~$2.03 | ~2.0% |
| Tempo | $0.001 | $0 | — | $2.00 | ~$2.00 | ~2.0% |
At $100, the off-ramp dominates. Network-layer differences between the cheap chains are lost in the rounding. The meaningful split is between chains where network fees are below $0.10 (Solana, Base, Tempo, Tron-staked) and Ethereum or Tron-unstaked, where the network fee itself is material.
The UN's Sustainable Development Goal targets remittance costs below 3%. On Ethereum or Tron without staking, a $100 transfer is already at or above that threshold from network fees alone before the off-ramp.
Scenario 2: $1,000 business payment (B2B invoice)
A company pays a contractor invoice of $1,000. Both parties use self-custody wallets; no off-ramp (the contractor holds stablecoins).
| Chain | Network fee | Gas-token cost | PSP markup | Total | % of transfer |
|---|---|---|---|---|---|
| Ethereum | $2.00 (mid) | $0.02 | — | ~$2.02 | ~0.20% |
| Tron (pre-staked) | ~$0 | (capital cost) | — | < $0.01 | < 0.001% |
| Solana | $0.001 | ~$0 | — | ~$0.001 | < 0.001% |
| Base | $0.02 (mid) | $0.01 | — | ~$0.03 | ~0.003% |
| Tempo | $0.001 | $0 | — | ~$0.001 | < 0.001% |
At $1,000 with no off-ramp, Ethereum is still meaningful — $2 is 0.2% of the invoice value, which accumulates across many transfers. For a company processing 500 invoices per month at $1,000 each, Ethereum fees at $2 each run $1,000 per month — about $12,000 a year — in gas costs alone.
Solana, Tron (staked), and Tempo are all effectively zero at this size. The distinction shifts to operational considerations: does the business want to manage a TRX staking position? Does it want SOL in addition to stablecoins? Tempo's USD-only stack is simpler to operate, but the fee difference versus Solana at this transfer size is immaterial.
Scenario 3: $100,000 treasury settlement
A business moves $100,000 from its treasury to a foreign subsidiary's wallet. Both sides hold stablecoins; no immediate off-ramp. Settlement must be verifiably final for accounting purposes.
| Chain | Network fee | Gas-token cost | Finality time | Finality type | Total network cost |
|---|---|---|---|---|---|
| Ethereum | $2–$5 | $0.05 | 12–15 min | Probabilistic economic | ~$5 |
| Tron (pre-staked) | ~$0 | (capital cost) | ~57 sec | Probabilistic (solidified) | < $1 |
| Solana | $0.001 | ~$0 | 1–2 sec (optimistic) | Optimistic | < $0.01 |
| Base | $0.02–$0.05 | $0.01 | Soft: 1–5 sec; Hard: ~1 day | Two-tier | < $0.10 |
| Tempo | $0.001 | $0 | < 1 sec | Deterministic (BFT) | < $0.01 |
At $100,000, the absolute network fee matters less — even Ethereum's $5 is 0.005% of the transfer value. What matters more is finality quality.
For a treasury settlement that needs to appear as a finalised line item in accounting systems, "probabilistic" is a complication. Ethereum's 12–15-minute window is manageable but means the transaction cannot be recorded as settled for a quarter-hour. Tron's 57-second window is shorter. Base's hard finality takes about a day.
Tempo provides deterministic finality in under a second: the transfer is irreversibly settled, recorded as final, no confirmation count required, no re-org risk window. For treasury operations that run thousands of settlements per month, automated reconciliation against deterministically final transactions eliminates an entire class of exception handling.
Stripe's published rate for stablecoin settlement on Tempo is $0.001 per transaction — flat, regardless of dollar amount. At $100,000, that is effectively zero as a percentage.
The off-ramp question
For transfers where the recipient converts back to local currency, the off-ramp fee is typically the dominant cost. Off-ramp coverage varies by chain:
Ethereum and Tron have the most mature off-ramp ecosystems — virtually every exchange and off-ramp provider supports both. This matters most for users in emerging markets where USDT-TRC20 is the default stablecoin format.
Solana has strong exchange coverage (most major exchanges support Solana USDC and USDT withdrawals directly), and Circle's native USDC issuance on Solana has expanded institutional coverage.
Base has Coinbase's off-ramp as a native integration — Coinbase users can move USDC from Base directly to their Coinbase account and cash out. This is a significant coverage advantage for US retail.
Tempo launched in March 2026. Off-ramp coverage is building: MoneyGram (200+ countries) joined as an anchor remittance validator and settlement partner, and Flutterwave covers 34 African markets for Send App transfers. This is meaningful coverage for key corridors, but not yet the breadth of Tron or Ethereum.
For a $100 remittance to a corridor where Tempo has off-ramp coverage, the total cost is competitive. For a corridor not yet covered, the routing is indirect.
Summary by transfer size and use case
| Transfer | Dominant cost | Best-value chains | Key consideration |
|---|---|---|---|
| $100 remittance | Off-ramp | Solana, Base, Tempo, Tron (staked) | Off-ramp corridor coverage matters most |
| $1,000 B2B, no off-ramp | Negligible | Solana, Tempo, Tron (staked) | Operational simplicity (gas token or not) |
| $100,000 treasury | Finality quality | Tempo, Solana (optimistic), Base (soft) | Deterministic vs probabilistic finality for accounting |
| High volume (1,000+ transfers/day) | Network fee × volume + gas mgmt overhead | Tempo, Solana | No gas-token overhead; Tempo has Payment Lanes for blockspace guarantee |
The bottom line
The cheapest chain at the network layer is not always the cheapest end-to-end. For small consumer remittances, the off-ramp fee swamps every other variable — the right chain is the one with the best corridor coverage, which today means Tron or Ethereum for most markets, with Solana and Base strong in US and USDC-heavy corridors, and Tempo building coverage fast with MoneyGram and Flutterwave as anchors.
For business payments and high-volume transfers, the gas-token management overhead and finality quality become the differentiating factors. Ethereum is expensive at volume. Tron's staking model works but requires managing a TRX position. Solana is cheap but asks for SOL. Tempo is the only chain where the fee is in dollars, the finality is deterministic, and the operational stack stays dollar-denominated throughout.
For the full side-by-side on fees, finality, and TPS, see blockchains for stablecoin payments, compared. For the mechanics of gas-token friction, see why does gas matter for stablecoin payments?.