Tools/Explorers

Bitcoin Transaction Accelerators Compared

Compare Bitcoin transaction accelerator methods: RBF, CPFP, mining pool accelerators, and third-party services for unsticking low-fee unconfirmed transactions.

Spark TeamInvalid Date

Why Bitcoin Transactions Get Stuck

A Bitcoin transaction becomes "stuck" when its fee rate is too low to compete for block space. Every Bitcoin node maintains a local mempool: a waiting area for unconfirmed transactions where they compete for inclusion based on fee rate (measured in sat/vB). When demand for block space exceeds the ~4 MB weight limit per block, miners prioritize transactions offering the highest fees, leaving low-fee transactions waiting for hours, days, or until they expire from the mempool entirely.

The default Bitcoin Core mempool holds 300 MB of transaction data. When this limit is reached, the node evicts the lowest-fee transactions first and raises its minimum acceptance fee dynamically. During congestion spikes, this effective minimum can climb well above the static floor. Transactions below the cutoff are dropped and must be rebroadcast once fees decline. For a deeper look at these dynamics, see our research on how Bitcoin's fee market works.

Common causes of stuck transactions include: fee estimators underestimating demand during rapid congestion increases, wallets using static fee defaults, transactions broadcast just before a surge in network activity (such as a new token launch or protocol event), and consolidation transactions sent at rock-bottom rates during quiet periods that get caught by a sudden fee spike.

Acceleration Methods Overview

There are four categories of approaches to accelerate a stuck Bitcoin transaction, ranging from self-service protocol-level techniques to paid third-party services. The right choice depends on whether you are the sender or recipient, which wallet you used, and how urgently you need confirmation.

MethodWho Can Use ItCostSuccess RateSpeedRequirements
RBF (Replace-By-Fee)Sender onlyFee difference + relay overheadVery highNext block (if fee is sufficient)Wallet with RBF support
CPFP (Child-Pays-For-Parent)Sender or recipientChild tx fee covers both parent and childHighNext block (if combined rate is sufficient)Spendable output from stuck tx
Mining pool acceleratorAnyone with the TXIDFree (limited) to $25-200+Moderate (depends on pool hashrate)Minutes to hoursAccount or payment
Third-party acceleratorAnyone with the TXID$5-500+ (varies widely)Variable (verify legitimacy)Minutes to hoursPayment (Lightning, credit card)

For a focused comparison of just the two protocol-level methods, see our RBF vs CPFP comparison tool.

RBF: Replace-By-Fee

Replace-By-Fee (RBF) lets the sender broadcast a replacement transaction that spends the same inputs but pays a higher fee. Under BIP 125, the original transaction signals replaceability by setting at least one input's nSequence value below 0xFFFFFFFE. Nodes running Bitcoin Core accept the replacement if it meets several rules: it must spend at least one of the same inputs, pay a strictly higher absolute fee, cover the minimum relay fee for its own size, and not evict more than 100 transactions from the mempool.

Bitcoin Core v28.0 (October 2024) changed the default to mempoolfullrbf=1, and v29.0 (April 2025) removed the option entirely, making full RBF mandatory and non-configurable. Any unconfirmed transaction can now be replaced regardless of whether it signaled BIP 125 replaceability. The practical effect is that zero-confirmation on-chain payments are no longer considered reliable.

RBF is the most efficient acceleration method because the replacement transaction does not add block weight: it replaces the original entirely. According to Bitcoin Core's documentation, RBF is 30% to 90% more fee-efficient than CPFP. The only cost overhead is the fee difference plus a small amount to cover relay of the replacement (at least the replacement's size multiplied by the minimum relay fee rate). Most modern wallets support RBF natively, making it the recommended first option for any sender with a stuck transaction.

CPFP: Child-Pays-For-Parent

Child-Pays-For-Parent (CPFP) works by creating a new "child" transaction that spends an output from the stuck "parent." The child pays a fee high enough that the combined fee rate of both transactions (total fees divided by total virtual size) exceeds the threshold for block inclusion. Since Bitcoin Core v0.13.0, miners evaluate transaction packages by ancestor fee rate when assembling block templates, so a high-fee child pulls its low-fee parent into the block.

CPFP has a key advantage over RBF: the recipient can use it. If you are waiting to receive bitcoin and the sender's transaction is stuck, you can spend your unconfirmed output with a high-fee child to accelerate confirmation. The sender can also use CPFP by spending a change output, though RBF is usually more efficient in that case.

The tradeoff is cost: CPFP adds a second transaction to the block, so the child's fee must compensate for both its own weight and the parent's weight at the target fee rate. For example, if a 300 vB parent paid 5 sat/vB and you want a combined rate of 25 sat/vB, a 150 vB child must pay 9,750 sats (65 sat/vB individually) to achieve that package rate. Bitcoin Core v31.0 (April 2026) replaced the legacy ancestor and descendant limits with a cluster mempool design, where clusters are limited to 64 transactions or 101 kB virtual size.

Bitcoin Core v28.0 introduced TRUC (Topologically Restricted Until Confirmation) transactions (BIP 431), also called v3 transactions, which restrict package topology to one parent and one child (with the child limited to 1,000 vB). This makes CPFP-based fee bumping more reliable by preventing transaction pinning attacks. The same release added 1-parent-1-child package relay support at the P2P layer, and v31.0 extended this to non-TRUC transactions.

For a complete guide to using both RBF and CPFP in practice, see our RBF and CPFP fee bumping guide.

Mining Pool Accelerators

Mining pool accelerators work by having the pool operator add your transaction to their block template with higher priority than its fee rate would normally warrant. The effectiveness depends on the pool's share of total network hashrate: a pool with 10% of hashrate will find roughly 10% of blocks, so there is a probabilistic delay before they mine the block that includes your transaction.

Mempool.space Accelerator

Mempool.space operates one of the most transparent acceleration services available. When viewing a pending transaction on mempool.space, users can pay an out-of-band fee (via Lightning, Cash App Pay, Apple Pay, Google Pay, or preloaded credit) to have the transaction prioritized by partner mining pools including Foundry USA Pool, MARA Pool, SBI Crypto, SpiderPool, and Ocean. The fee is calculated based on the delta between the transaction's current fee rate and the rate needed for next-block confirmation, plus a service margin.

ViaBTC Accelerator

ViaBTC offers both free and paid acceleration tiers. The free tier allows submitting a TXID for prioritization in the next ViaBTC-mined block, but slots are limited (roughly 100 per hour) and fill up quickly during congestion. The paid tier charges a variable fee based on transaction size and current conditions, with reported costs ranging from approximately $20 to $200+ during high-congestion periods.

MARA Slipstream

Marathon Digital's Slipstream service accepts transaction submissions directly, bypassing the standard P2P mempool relay. While originally designed for non-standard transactions (such as large Ordinals inscriptions), it also functions as a transaction accelerator. Slipstream is useful for transactions that nodes might not relay due to policy restrictions rather than just low fees.

Warning: Many websites advertising "Bitcoin transaction acceleration" are scams that charge fees without any mining pool partnerships. Verify that any accelerator has a direct relationship with active mining pools before paying.

Wallet Support for Fee Bumping

Not every wallet supports RBF and CPFP equally. The following table summarizes fee bumping capabilities across popular Bitcoin wallets.

WalletRBF Send (signals by default)RBF BumpCPFP SupportFee Estimation
Bitcoin CoreYes (default since v24.0)YesYes (manual via raw tx)Built-in (conservative and economical modes)
ElectrumYes (default)YesYes (right-click menu)Built-in
SparrowYes (default)YesYes (since v1.4.0)Built-in with mempool.space integration
BlueWalletYesYesYes (Native SegWit only)Built-in
Ledger LiveYesYesYesBuilt-in
Trezor SuiteYes (default)YesYes (via Coin Control)Built-in
WasabiYes (default)YesYes (since v2.0.4)Built-in
NunchukYesYesYesBuilt-in

If your wallet does not support RBF bumping, you may still be able to use CPFP if you can spend an unconfirmed output. Check whether your transaction signaled RBF using our RBF detector tool.

Decision Tree: Choosing the Right Method

Use this decision framework to determine which acceleration approach fits your situation:

Are you the sender or the recipient?

  • Sender with a wallet that supports RBF: use RBF. It is the cheapest and most reliable method. Broadcast a replacement with a higher fee rate targeting the current estimated fee for your desired confirmation speed.
  • Sender without RBF wallet support but with a change output: use CPFP on your change output. Since full RBF is now mandatory in Bitcoin Core v29.0+, you can also create an RBF replacement even if the original did not signal it.
  • Recipient waiting for an incoming payment: use CPFP by spending your unconfirmed output with a high-fee child transaction.
  • No spendable outputs and no RBF capability: use a mining pool accelerator (mempool.space is the most accessible option) or wait for congestion to clear.

How urgent is confirmation?

  • Within the next block: RBF or CPFP with a fee rate matching the top of the current mempool. Mining pool accelerators may take several blocks depending on the pool's hashrate share.
  • Within a few hours: any method works. RBF or CPFP with a moderate fee bump, or a mining pool accelerator.
  • No rush: wait. If the mempool clears (common during weekends and off-peak hours), your transaction will confirm at its original fee rate. Most transactions are evicted from the mempool after 14 days if unconfirmed, at which point you can rebroadcast with a higher fee.

Preventing Stuck Transactions

The best acceleration strategy is prevention. These practices minimize the risk of stuck transactions:

  • Use dynamic fee estimation: choose a wallet that pulls real-time fee data from the mempool rather than using static defaults. Our Bitcoin fee estimator shows current rates for different confirmation targets.
  • Always signal RBF: ensure your wallet creates transactions with RBF signaling enabled so you can bump fees if needed. Most modern wallets do this by default.
  • Consolidate UTXOs during low-fee periods: reducing the number of inputs in future transactions lowers their size and cost. See our UTXO consolidation guide for best practices.
  • Batch transactions when possible: combining multiple payments into one transaction reduces total block weight and cost per payment.
  • Use SegWit addresses: transactions spending from bech32 (bc1q) or Taproot (bc1p) addresses are smaller in virtual bytes, meaning lower fees for the same fee rate.

Avoiding On-Chain Congestion Entirely

On-chain fee volatility is a fundamental consequence of Bitcoin's limited block space. Layer 2 protocols offer an alternative by moving transactions off the main chain while preserving Bitcoin's security guarantees.

The Lightning Network processes payments through pre-funded payment channels, avoiding the mempool entirely for routine transfers. Payments settle in seconds regardless of on-chain congestion. However, opening and closing channels still requires on-chain transactions, and channel management introduces its own complexity. For a broader view of how these scaling approaches compare, see our research on mempool congestion economics.

Spark takes a different approach as a Bitcoin Layer 2: it enables instant, near-zero-fee transfers of both bitcoin and stablecoins like USDB without requiring channel management or on-chain transactions for everyday use. For users who primarily need fast, reliable transfers, L2 solutions eliminate the entire category of stuck-transaction problems.

Frequently Asked Questions

How long does an unconfirmed Bitcoin transaction take to expire?

By default, Bitcoin Core nodes drop unconfirmed transactions from their mempool after 336 hours (14 days). However, if any node on the network still has the transaction in its mempool and rebroadcasts it, the timer resets. In practice, very low-fee transactions during sustained congestion may be evicted much sooner as the mempool fills and the minimum fee threshold rises. Once a transaction is fully evicted from the network, the funds return to the sender's available balance and can be spent in a new transaction.

Can I cancel a Bitcoin transaction that is stuck?

You cannot cancel a Bitcoin transaction in the traditional sense, but you can effectively cancel it using RBF. Create a replacement transaction that sends the same inputs back to your own address with a higher fee. Since full RBF is mandatory in Bitcoin Core v29.0+, this works regardless of whether the original transaction signaled replaceability. The replacement confirms in place of the original, effectively canceling the payment. This only works while the transaction is unconfirmed: once a transaction has at least one confirmation, it cannot be reversed.

Are Bitcoin transaction accelerator services safe?

Legitimate accelerators operated by established mining pools or well-known services (such as mempool.space or ViaBTC) are generally safe to use. However, many websites advertising "Bitcoin transaction acceleration" are scams that charge fees without any mining pool partnerships or ability to actually prioritize your transaction. Before using any accelerator, verify that the service operator has a direct relationship with active mining pools. When possible, prefer protocol-level methods (RBF or CPFP) over third-party services, as they do not require trusting an intermediary.

Does RBF work if the original transaction did not signal it?

Yes. Bitcoin Core v28.0 (October 2024) made full RBF the default, and v29.0 (April 2025) made it mandatory by removing the configuration option entirely. All Bitcoin Core nodes now accept replacement transactions regardless of BIP 125 signaling. Since the vast majority of the network runs Bitcoin Core, a properly constructed replacement transaction will propagate and be included by miners. You can verify whether your transaction signals RBF using our RBF detector, though signaling is no longer required for replacement.

What is the cheapest way to unstick a Bitcoin transaction?

RBF is the cheapest method because the replacement transaction replaces the original without adding extra block weight. You pay only the difference between the original fee and the new fee (plus a small relay overhead). CPFP is more expensive because the child transaction adds weight that must also be paid for. Mining pool accelerators charge additional service fees on top of the fee deficit. The cheapest option of all is simply to wait: if the mempool clears during off-peak hours, your transaction may confirm at its original fee rate without any intervention.

Can the recipient speed up a Bitcoin transaction?

Yes, using CPFP. The recipient creates a child transaction that spends their unconfirmed output from the stuck transaction, paying a fee high enough that the combined package fee rate qualifies for block inclusion. This is the only protocol-level method available to recipients, since RBF requires access to the original transaction's private keys (which only the sender has). Alternatively, the recipient can submit the transaction ID to a mining pool accelerator.

How much does it cost to accelerate a Bitcoin transaction?

Costs vary dramatically based on fee market conditions. During low congestion, an RBF bump might cost a few hundred satoshis (under $1). During fee spikes, bumping a transaction to the front of the mempool can cost $10-50+ depending on the transaction's virtual size and how far below the current market rate it sits. Mining pool accelerators add service fees on top: ViaBTC's paid tier ranges from approximately $20-200+, while mempool.space's pricing is based on the fee delta plus a margin. The cost fundamentally scales with the gap between your transaction's fee rate and the current rate needed for confirmation.

This tool is for informational purposes only and does not constitute financial advice. Accelerator service availability, pricing, and effectiveness change frequently. Mining pool hashrate shares fluctuate, affecting accelerator reliability. Always verify current service status and fees before using any third-party accelerator. When possible, use protocol-level methods (RBF, CPFP) over third-party services.

Build with Spark

Integrate bitcoin, Lightning, and stablecoins into your app with a few lines of code.

Read the docs →