Seven of the largest Bitcoin mining pools — Foundry, AntPool, F2Pool, SpiderPool, MARA Pool, Block Inc., and DMND — joined the Stratum V2 Working Group on May 7, 2026, consolidating approximately 75% of the network's ~998 EH/s hashrate behind an open protocol that transfers block template constru...
"We are proud to support the broader adoption of Stratum V2. Aligning around an open, interoperable standard lets the industry collaborate on efficiency, security, and decentralization." — Andy Zhou, CEO, AntPool
Seven of the largest Bitcoin mining pools — Foundry, AntPool, F2Pool, SpiderPool, MARA Pool, Block Inc., and DMND — joined the Stratum V2 Working Group on May 7, 2026, consolidating approximately 75% of the network's ~998 EH/s hashrate behind an open protocol that transfers block template construction from pool operators to individual miners. Foundry alone accounts for 34.2% of global hashrate; AntPool adds 14.2%; F2Pool contributes 11.3%.
The move addresses a structural vulnerability that has shadowed Bitcoin for years: fewer than six pool operators have controlled transaction selection for over 95% of mined blocks. Stratum V2's Job Declaration sub-protocol lets miners choose which transactions enter their blocks, stripping pool operators of a power that critics say creates a single point of censorship. Two pools — Braiins Pool and DMND — already run Stratum V2 in production. The working group, co-founded in 2022 by Braiins and Spiral (Block Inc.'s Bitcoin development arm), projects that V2 could become the default protocol for new ASIC firmware shipments by end of 2026, potentially reaching 40–60% of live network hashrate.
Bitcoin's mining pool landscape has consolidated steadily since the 2024 halving. According to data from miningpool.observer, six pools produce more than 95% of all blocks. Foundry and AntPool alone account for roughly 48% of total hashrate. This concentration means two corporate entities — Digital Currency Group (Foundry's parent) and Bitmain (AntPool's operator) — effectively decide which transactions enter nearly half of all Bitcoin blocks.
The architecture responsible is Stratum V1, a protocol released in 2012 by Marek "Slush" Palatinus. Under V1, pool operators build block templates and push them to connected miners. Miners receive a template, hash against it, and return valid proofs-of-work. They have no say in transaction selection. The protocol transmits data in plaintext with no encryption.
This design has produced documented censorship events. In May 2021, Marathon Digital launched an explicitly "OFAC-compliant" mining pool that filtered transactions associated with sanctioned addresses. F2Pool was observed censoring OFAC-sanctioned transactions in late 2023. While both pools later reversed course under community pressure, the incidents demonstrated that transaction filtering at the pool level is not theoretical — it is operationally trivial.
The concern is not that any single pool will permanently censor transactions. It is that the current architecture makes censorship a policy decision rather than a technical impossibility. Two or three pools coordinating — whether voluntarily or under regulatory compulsion — could delay confirmation of targeted transactions indefinitely.
Stratum V2 was specified in 2019 by Pavel Moravec and Jan Čapek of Braiins, in collaboration with Matt Corallo, then of Chaincode Labs (now Spiral). Corallo had proposed Betterhash in 2018 as a first attempt to return transaction selection to miners. Stratum V2 absorbed and expanded that proposal.
The protocol introduces three sub-protocols:
1. Mining Protocol. Handles the core communication between miners and pools. Unlike V1, all data is end-to-end encrypted using the Noise Framework, preventing man-in-the-middle attacks and eliminating plaintext exposure of mining data.
2. Job Declaration Protocol. This is the critical component. It allows individual miners to construct their own block templates — selecting transactions from their local mempool — and submit them to the pool for validation. The pool retains responsibility for reward distribution. The miner gains authority over transaction selection.
3. Template Distribution Protocol. Provides miners with the information needed to build valid block templates, sourced from their own Bitcoin Core node.
Bandwidth improvements are significant: approximately 60% reduction on the pool side and 70% reduction for miners, according to Braiins. The SRI (Stratum V2 Reference Implementation) reached v1.0 in March 2024 and has since shipped through v1.6.0. The working group has split the codebase into library and application repositories, with precompiled binaries and Docker images available for deployment.
Auradine has shipped the first ASIC miner with native Stratum V2 support, removing the firmware compatibility barrier for new hardware deployments.
The seven entities that joined the working group on May 7:
| Entity | Estimated Hashrate Share | Role | |--------|--------------------------|------| | Foundry | 34.2% | Mining pool (DCG subsidiary) | | AntPool | 14.2% | Mining pool (Bitmain-operated) | | F2Pool | 11.3% | Mining pool | | SpiderPool | 10.5% | Mining pool | | MARA Pool | 4.7% | Mining pool (Marathon Digital) | | Block Inc. | N/A (hardware) | ASIC manufacturer, Spiral parent | | DMND | ~1% | Mining pool (already running V2) |
Combined, the pool operators represent approximately 75% of Bitcoin's network hashrate. Block Inc.'s participation is notable because it manufactures Bitcoin mining hardware — its involvement signals that future Block-designed ASICs will likely ship with native V2 support.
The composition also matters for what it implies about competitive dynamics. These pools are direct competitors. Their collective endorsement of an open standard suggests the perceived economic and reputational benefits of Stratum V2 outweigh competitive concerns about standardization.
It is worth noting what joining the working group does not mean: it does not mean these pools have deployed Stratum V2 in production. Only Braiins Pool and DMND currently run V2 on their production infrastructure. The remaining five have committed to integration and testing timelines that remain unspecified.
The timing of this announcement is not coincidental. Bitcoin mining economics are under severe pressure.
According to CoinShares' Q1 2026 mining report, hashprice closed Q1 2026 at approximately $23.9 per PH/s per day — the lowest reading since 2018 and roughly 80% below the pre-halving peak of ~$120 per PH/s per day. Production costs for mid-tier U.S. public miners have climbed to $45,000–$60,000 per BTC. As of May 11, hashprice stands at $38.57 per PH/s per day, near breakeven for mid-generation hardware. Network difficulty is expected to rise from 132.47T to 135.64T on May 15.
Approximately 20% of Bitcoin miners are currently operating at a loss.
In this environment, the Stratum V2 working group's claim of up to 7.4% higher profitability through lower latency and better fee capture is economically material. For a miner operating at breakeven, a 7.4% efficiency gain is the difference between survival and shutdown. The gains come from three sources:
For publicly traded miners like Marathon Digital (which operates MARA Pool), even marginal improvements in unit economics are reportable metrics that affect equity valuations.
Stratum V2 is not the only attempt to decentralize block template construction. OCEAN, a mining pool founded in November 2023 by Bitcoin Core developer Luke Dashjr with a $6.2 million seed round led by Jack Dorsey, operates its own protocol called DATUM (Decentralized Alternative Templates for Universal Mining).
DATUM shares Stratum V2's core objective — miners build their own blocks — but differs in implementation. OCEAN's model is non-custodial: coinbase payouts go directly to miners without pool custody of funds. DATUM charges a 2% fee (1% for miners using the DATUM protocol). OCEAN commands roughly 2% of global Bitcoin hashrate as of early 2026.
In April 2025, Tether announced it would deploy hashrate on OCEAN, providing institutional validation for the decentralized mining thesis.
The two protocols are not directly interoperable. A miner must choose between a Stratum V2-compatible pool and OCEAN's DATUM infrastructure. The working group's consolidation of 75% of hashrate behind V2 creates significant network effects that DATUM, at 2% hashrate, cannot currently match. Whether DATUM remains a viable alternative or is absorbed into the V2 ecosystem is an open question.
The Stratum V2 working group projects the following milestones:
Several risks merit attention:
Adoption lag. Joining a working group is not the same as deploying production infrastructure. Pool operators may delay implementation if the engineering cost exceeds short-term revenue benefits.
Firmware fragmentation. Older ASICs without native V2 support will require firmware updates or proxy solutions. The installed base of V1-only hardware is substantial.
Regulatory pressure. The entire premise of Stratum V2 — that miners, not pools, should select transactions — may attract regulatory attention. If regulators mandate that pools enforce transaction filtering (as OFAC compliance could require), pools may resist enabling the Job Declaration protocol even while adopting V2's other features.
Miner sophistication. Running a full Bitcoin node and constructing valid block templates requires technical competence that not all mining operations possess. Default configurations may still defer to pool templates, undermining the decentralization thesis.
The consolidation of 75% of Bitcoin hashrate behind Stratum V2 represents the most significant structural shift in Bitcoin mining governance since the emergence of pooled mining in 2010. The protocol directly addresses the centralization of transaction selection — a vulnerability that has been discussed for years but never addressed at scale.
The economic context matters. Miners are not adopting V2 out of ideological commitment to decentralization. They are adopting it because compressed margins make the protocol's efficiency gains existentially relevant. The 7.4% profitability improvement is a survival metric, not a feature bullet point.
Whether this translates to actual deployment depends on execution. Working group membership is a statement of intent, not a guarantee of production infrastructure. The gap between commitment and deployment will determine whether 2026 is remembered as the year Bitcoin mining decentralized its most centralized function, or as another year of protocol proposals that failed to reach production.