Espresso Reports 400-Millisecond Finality After Two Years on Mainnet

By Tom Whitfield

Espresso says its mainnet has produced about 31.7 million blocks in two years while reducing finality from roughly 10 seconds at launch to about 400 milliseconds. The project also reports that stated data throughput rose to 181 MB/s from 30 MB/s.

Both figures come from Espresso, which describes the protocol as a “base layer for other blockchains.” Its architecture includes neither a virtual machine nor smart contracts, and Espresso attributes the performance gains to that streamlined design.

That omission defines what the measurements can support. Espresso presents a narrow base-layer system centered on data throughput and finality, not a general-purpose environment for executing smart contracts. The reported speed therefore describes this streamlined architecture rather than the wider set of operations a virtual machine might support.

How much stablecoin activity could the network handle?

Espresso says the 181 MB/s rate can support “one million stablecoin transactions per second.” The project derives that estimate from data volume and a simple token transfer of roughly 180 bytes. It is a capacity calculation, not a direct count of transactions completed each second on one chain.

MB/s measures data moving through the system; transactions per second measures completed operations. The conversion depends on a chosen transaction size, which is why Espresso cites a roughly 180-byte simple transfer. The estimate indicates potential capacity under those assumptions, not observed stablecoin activity.

• Reported mainnet data throughput: 181 MB/s, up from 30 MB/s.
• Capacity estimate: 1 million simple stablecoin transfers per second at roughly 180 bytes each.
• Reported mainnet finality: about 400 milliseconds, down from about 10 seconds at launch.

The qualification prevents a direct capacity claim. A one-million-transfer second would require the live chain to process that volume in that interval. Espresso has not reported such a live-mainnet count, and its 181 MB/s figure is described as stated throughput rather than a completed transaction rate.

What did the separate benchmark measure?

A separate benchmark measured 181 MB/s across 100 validators while emulating network delays between global locations. At peak capacity, it recorded a median settlement time of 778 milliseconds. The test did not measure live-mainnet throughput.

The benchmark matched Espresso's stated data rate but recorded a settlement result nearly twice the roughly 400-millisecond mainnet finality figure. Because the test emulated geographic delays and did not use live-mainnet traffic, its 778-millisecond result cannot stand in for sustained production performance.

For chain operators, the figures answer different questions. The mainnet record covers 31.7 million blocks, two years of operation and a reported 400-millisecond finality result. The benchmark covers 100 validators, emulated global latency and 778 milliseconds of median settlement at peak capacity. Neither directly measures one million live stablecoin transfers in one second.

The transaction estimate uses a roughly 180-byte simple transfer, so it does not establish capacity for stablecoin operations with different record sizes. The separate benchmark tests data rate and settlement, not the one-million-transfer claim itself.

A live-mainnet exercise would need actual transaction counts, sustained throughput and finality under one workload before the capacity estimate becomes an operating benchmark. Espresso has not disclosed a deadline or schedule for that validation.

For stablecoin infrastructure, the next market-structure checkpoint is a reproducible 181 MB/s run that reports actual transfers and finality under one load. Until then, Espresso has a two-year block-production record and a separate emulated capacity test.