What P2Pool was
P2Pool was a decentralized Bitcoin mining pool created by Forrest Voight in 2011. Instead of a central pool operator coordinating miners and distributing payouts, P2Pool ran its own peer-to-peer share chain, and paid every contributing miner directly inside each Bitcoin coinbase transaction. There was no operator to trust, no operator to seize funds, no operator to censor transactions.
For its time, it was a remarkable design — and the spiritual ancestor of every project listed below.
Why it stopped being viable
Three structural problems made original P2Pool unworkable for small Bitcoin miners:
- Variance. P2Pool's share chain ran at a higher effective difficulty than centralized pools, so a hobbyist contributing a few TH/s could go weeks without recording a share.
- Coinbase bloat. Paying every miner directly inside the coinbase grew the generation transaction. A typical block paid a few hundred addresses, and the largest (block 236,962, in May 2013) paid 553 in a 19 kB coinbase. That size made P2Pool blocks less competitive to mine, eating into rewards.
- Dust economics. As Bitcoin transaction fees climbed, payouts to small miners became smaller than the cost of ever spending the resulting UTXO.
The original P2Pool repository (github.com/forrestv/p2pool) has been effectively dormant since the late 2010s. The network itself has not quite stopped. It still finds about one Bitcoin block a year: seven between the start of 2020 and June 2026, the last of them paying 52 addresses.
Each of these problems has a modern answer being actively worked on. Uncle blocks reduce variance. Selective coinbase payouts to large miners control bloat. Atomic-swap markets clear the dust at the edges. The projects below are how the design space has moved on.
The state of decentralized mining in 2026
The conversation today is less "rebuild P2Pool" and more "decompose what a pool actually does":
- Template construction — who decides which transactions go in a block? Stratum V2 and DATUM push this from pools back to individual miners.
- Share accounting — how is each miner's work measured? P2Pool v2 uses uncle blocks; Braidpool uses a DAG-based share chain.
- Payout custody — who holds the coins between block-found and miner-paid? OCEAN pays miners straight from the coinbase once they pass a payout threshold; Monero's P2Pool eliminates it.
- Small-miner liquidity — what do you do with payouts that are smaller than the fee to spend them? P2Pool v2's planned atomic-swap share market is the most novel answer in this space; market makers would take the dust-aggregation pain in exchange for the virgin-coin premium.
A fully decentralized Bitcoin pool, in the original P2Pool spirit, requires progress on all four. As of November 2026 no production system delivers all four for Bitcoin, but the pieces are coming together, and serious effort is going into each.
Projects worth knowing
Pool protocol · Live on testnet4 · The direct reboot
A from-scratch reboot of P2Pool for Bitcoin, led by Kulpreet Singh. Written in Rust on libp2p and rust-bitcoin. Three design moves directly attack the problems that killed the original: uncle shares (a share can reference up to three, and an uncle's work still counts and is paid at a small discount) so shares that arrive late or race another are not orphaned and variance drops; direct coinbase payouts for the top N miners (up to ~500 outputs) so the largest contributors get paid non-custodially without the coinbase growing with every miner who joins; and an atomic-swap market on the share chain so smaller miners can sell their shares to market makers for immediate liquid Bitcoin, sidestepping the dust problem without custody or covenants. As of November 2026 the share chain, its accounting and the coinbase payouts run on testnet4 and the swap market is still being built. Not yet a mainnet pool, but the project most directly aligned with reviving the original P2Pool thesis.
Protocol · Active deployment
Binary, encrypted successor to the Stratum mining protocol that has been used since 2012. Its critical contribution is letting individual miners — not pools — construct block templates. In May 2026 Foundry, AntPool, F2Pool, SpiderPool and MARA, pools that together find about three quarters of Bitcoin's blocks, joined the Stratum V2 working group, along with Block and DMND. Joining is not deploying: the pools the project lists as running it in production (Braiins Pool, DMND, Public Pool and a few solo pools) are still a small share of hashrate. Reference implementation: github.com/stratum-mining.
Pool · Live · Non-custodial, DATUM-enabled
A Bitcoin mining pool that describes itself as non-custodial: miners are paid directly in the block's coinbase once their earnings pass a payout threshold, and smaller balances accrue or can go out over Lightning. It uses DATUM (Decentralized Alternative Templates for Universal Mining) to let participating miners build their own block templates. Transparent payout accounting via TIDES. The closest thing today to a "production decentralized pool" for Bitcoin.
Pool protocol · Active development
An ambitious from-scratch decentralized pool design built around a DAG-based share chain that settles to Bitcoin at the end of each difficulty epoch (~2 weeks) via a threshold signature among recent block finders. Each miner runs its own Bitcoin node and constructs its own templates. The closest spiritual successor to P2Pool, with modern thinking on share accounting and payout. Under active development; not yet in production.
Pool · Live · Built for Stratum V2
The first Bitcoin pool architected from day one for Stratum V2. End-to-end encryption against hashrate hijacking, and a transparent "SLICE" payout system (PPLNS adapted to miner-built templates). On 25 June 2026 it mined block 955,318, which it reports as the first known Bitcoin block built by a miner through Stratum V2 Job Declaration. Operator-run and VC-backed, and the pool still handles payouts, so less radical than Braidpool, but a meaningful step toward Stratum V2 at production scale.
Solo pool · Live · Bitaxe-friendly, zero-fee
A community-run, open-source solo mining coordinator with zero fees and no registration, by Benjamin Wilson. The default destination for Bitaxe and other open-source small miners, and self-hostable on Umbrel or Start9. In solo mode, if your miner finds a block the whole reward goes to your address. The site now also offers a PPLNS mode that splits rewards, plus Stratum V2 and DATUM endpoints. The most accessible on-ramp for hobbyist miners who want sovereignty over centralized convenience.
Pool · Live · The P2Pool model, applied to a fee-friendly chain
SChernykh's adaptation of the P2Pool model to Monero — the chain it actually fits on. Monero's small fees, short block times, and dynamic block size avoid the dust, variance, and coinbase-bloat problems that killed Bitcoin P2Pool. Worth studying as a working production example of the original P2Pool thesis.