• XRP Ledger activated amendment BatchV1_1 on mainnet on October 9, 2026, bundling up to 8 transactions.
• Batch supports four execution modes, including all-or-nothing and independent processing of bundled transactions.
• XRPL.org's 2017 commodity-hardware test recorded sustained throughput of 1,500 transactions per second.
Batch Amendment Live on Mainnet
Up to eight transactions can now clear through the XRP Ledger as one coordinated unit. The Batch feature went live on mainnet on October 9, 2026 under amendment BatchV1_1, letting an account submit between two and eight transactions that the network executes as a single workflow. XRPL.org's security report records the mainnet activation, and the official documentation lists the four execution modes the feature supports. The mode set defines how a bundle behaves. One mode applies the entire bundle only when every transaction inside it succeeds. A second applies only the first successful transaction, failing or skipping the rest. A third keeps applying transactions until the first failure appears. A fourth processes each transaction in the bundle independently. Under all four, the ledger records every applied transaction on its own line, so batching coordinates execution rather than compressing several entries into one. Failed or skipped members of a bundle leave no partial state behind, which removes a class of sequencing risk from multi-step applications. That distinction points to the intended use. Multi-step flows in which a later step depends on an earlier step's success can now run in a fixed order under explicit conditions. An asset transfer paired with its payment has been put forward as the reference case, since the transfer should not finalize unless settlement does. Submitters chain dependent steps into a single submission instead of firing separate transactions and hoping each lands in sequence. BatchV1_1 arrived through the ledger's standard amendment path, in which validators running the network's consensus mechanism signal support and an enabled amendment takes effect network-wide. The XRP Ledger is an open blockchain, and its documentation frames batch as a workflow tool, not a scaling claim. The activation has so far drawn little movement from the XRP price.
Throughput Depends on the Measurement
The one thing batching does not do is raise the ledger's per-second throughput. Each transaction inside a batch still settles on its own, so grouping changes the submission shape, not the processing load. Payment channels are a separate route for small payments altogether. A payer signs claims and exchanges them with the recipient outside the ledger, and the recipient settles the accumulated balance on the XRP Ledger afterward. Repeated micropayments therefore never touch mainnet one by one, which also means off-ledger claim volume cannot be counted as mainnet transactions per second. The Interledger protocol describes another layer: it passes payment packets between networks and adjusts balances between participants, often through payment channels. That too is a settlement arrangement, distinct from any figure measured on the ledger itself. XRPL.org's own benchmark dates to 2017, when commodity hardware of that era sustained 1,500 transactions per second in a general-purpose test. The number is a measured capacity result from that environment, not the mainnet's average transaction volume today. The comparisons readers reach for carry their own measurement problems. The annual report Visa filed with the U.S. Securities and Exchange Commission (SEC) as a public stock issuer lists 355.0 billion Visa-branded payment and cash transactions, of which 281.31 billion were processed directly by Visa. Dividing the directly processed count by 365 days yields roughly 8,921 transactions per second on average, an arithmetic conversion of annual volume rather than a disclosed throughput metric. Visa separately cites capacity to handle up to 83,000 transaction messages per second. A test result, an annual average and a message capacity describe different objects in different environments, and none converts cleanly into the others. Mastercard (MA) published a hackathon brief in which an autonomous AI agent, functioning as an AI crypto wallet with a fixed spending limit, pays a merchant and settles in Ripple's RLUSD stablecoin on the XRP Ledger. That is a test of a payment application, not a reading of ledger performance.
Where Batch Fits for Builders
COINOTAG's read is that Batch's value shows up in workflow design, not in headline throughput. The busiest settlement lane on the ledger is already stablecoins: our recent reporting put 92% of RLUSD swaps, worth $1.65 billion, running against XRP. Atomic multi-step flows give builders a way to tie those settlement conditions to delivery. Amendment traffic continues: the ledger's lending repayment fix still awaits the 80% validator vote that would spare borrowers an extra unit. Former Ripple CTO David Schwartz has put his own sway over the token's market price at very little, a fit for a quarter of protocol work over market moves. Up to eight transactions per batch now execute under one set of conditions, and each applied transaction still lands on the ledger as its own record.
