Seven of the largest Bitcoin mining pools — Foundry, AntPool, F2Pool, SpiderPool, Block Inc., MARA Foundation, and DMND — joined the Stratum V2 Working Group on May 7, 2026. Together, they represent approximately 75% of global Bitcoin hashrate. The move signals the most significant structural shi...
"Bitcoin's value proposition, I would argue, is that it is a censorship-resistant money." — Matt Corallo, Stratum V2 Co-Designer, Spiral/Block
Seven of the largest Bitcoin mining pools — Foundry, AntPool, F2Pool, SpiderPool, Block Inc., MARA Foundation, and DMND — joined the Stratum V2 Working Group on May 7, 2026. Together, they represent approximately 75% of global Bitcoin hashrate. The move signals the most significant structural shift in how Bitcoin blocks are constructed since the original Stratum protocol was introduced in 2012.
Stratum V2 decouples block template construction from hashrate aggregation. Under the current Stratum V1 standard, pool operators unilaterally decide which transactions enter each block. Stratum V2's Job Declaration sub-protocol allows individual miners to construct their own templates — selecting transactions from their own mempools — while still mining through a pool for reward aggregation. The protocol also introduces end-to-end encryption and reduces bandwidth consumption by 60-70%.
The adoption comes at a time when mining economics are under pressure. Hashprice sits at $38.57 per petahash per second per day, near breakeven for mid-generation hardware. Network difficulty is projected to rise from 132.47T to 135.64T on May 15. Stratum V2 proponents claim the protocol delivers up to 7.4% higher profitability through lower latency and better fee capture — a material margin in a post-halving environment where approximately 20% of operators face unprofitability.
Bitcoin mining pool concentration has reached levels that challenge the network's foundational design assumptions. As of May 2026, the hashrate distribution is as follows:
| Pool | Hashrate Share | |------|---------------| | Foundry USA | 34.2% | | AntPool | 14.2% | | F2Pool | 11.3% | | SpiderPool | 10.5% | | MARA Pool | 4.7% | | Others | ~25.1% |
Source: CoinDesk, Hashrate Index (May 2026)
Foundry and AntPool alone control approximately 48.4% of global hashrate. Four pools control roughly 70%. This means that four entities decide the transaction content of seven out of every ten Bitcoin blocks.
The concern is not purely theoretical. Under Stratum V1, the pool operator constructs the block template — deciding which transactions to include, in what order, and which to exclude. The miner receives this template and hashes against it with no input on block content. This architecture concentrates transaction selection power in a small number of operators, regardless of how distributed the underlying hashrate providers are.
According to analysis published by TFTC and referenced by Matt Corallo, several pools — including BTCcom, Binance Pool, Poolin, EMCD, and Rawpool — appear to share identical block templates and custom transaction prioritization with AntPool, suggesting coordinated template creation across nominally independent pools.
Network hashrate now sits at approximately 998 exahash per second, approaching the 1 ZH/s milestone first crossed in December 2025. The seven-day moving average fluctuates between 899 and 977 EH/s. Difficulty has been adjusted downward six times in 2026, with the most recent 2.3% reduction on May 1 at block height 947,520, bringing difficulty to 132.47 trillion.
Stratum V2 was originally designed by Jan Čapek and Pavel Moravec of Braiins, together with Matt Corallo (then at Chaincode Labs, now Spiral/Block). The protocol specification was released as an open standard, and the Stratum V2 Reference Implementation (SRI) v1.0 was published in March 2024, built in Rust with bindings for other languages.
The protocol operates through three sub-protocols:
Mining Protocol: Replaces V1's plaintext JSON-RPC communication with an encrypted binary format. This reduces pool-side bandwidth by approximately 60% and miner-side bandwidth by approximately 70%, according to the protocol specification. All miner-to-pool traffic is authenticated and encrypted, preventing man-in-the-middle attacks that could hijack hashrate by swapping pool credentials.
Job Declaration Protocol: This is the critical component. A miner running a Bitcoin full node constructs its own block template — selecting transactions from its local mempool based on its own criteria. The miner proposes this template to the pool. The pool verifies that the template is valid and that the miner's work attribution is correct, then allows the miner to hash against its own template. The pool's role reduces to reward aggregation and payout.
Template Distribution Protocol: Handles the communication between the Bitcoin node and the mining software for template construction and submission.
The net effect: Stratum V2 decouples the economic function of pools (aggregating hashrate and distributing rewards) from the governance function (deciding block contents). These two functions have been bundled since pooled mining began. Stratum V2 unbundles them.
The Stratum V2 Working Group was founded in 2022 by Braiins and Spiral (Block's Bitcoin technology arm). For four years, it operated as an independent open-source community with limited industry adoption. Braiins Pool and DMND were the only pools running Stratum V2 in production. DMND launched in 2025 as one of the first pools to offer full miner-selected templates.
On May 7, 2026, the Working Group announced seven new members:
The Working Group described the expansion as entering "a new phase of accelerated development and deployment." The group maintains Stratum V2 as a "public, vendor-neutral specification" to remove compatibility barriers across the mining ecosystem.
AntPool CEO Andy Zhou stated the company is "proud to support the broader adoption of Stratum V2," adding that aligning around an open, interoperable standard enables industry collaboration on efficiency, security, and decentralization. SpiderPool CTO Kenway Wang noted the protocol's support for miner-constructed templates is particularly useful for operators in bandwidth-constrained environments.
The group remains open to additional participants and has invited builders and operators across the mining ecosystem to contribute.
The adoption is not driven solely by ideology. Post-halving economics make every margin improvement material.
Hashprice: Currently at $38.57 per petahash per second per day, near breakeven for mid-generation ASIC hardware. According to CoinDesk, approximately 20% of mining operators currently face unprofitability.
Energy costs: As of March 2026, mining one Bitcoin in the United States requires more than 750,000 kWh at average business electricity rates of $0.1363/kWh, translating to approximately $106,135 in energy costs alone, according to BestBrokers research.
Stratum V2 efficiency gains: The protocol's proponents claim up to 7.4% higher profitability through three mechanisms:
In a market where hashprice hovers near breakeven, a 7.4% efficiency gain can mean the difference between operational viability and shutdown. This economic pressure is likely a stronger adoption driver than decentralization arguments alone.
Bitcoin's total daily electricity consumption in the United States stands at approximately 132.6 GWh, according to BestBrokers, corresponding to about 1.1% of total U.S. daily electricity demand. Any protocol improvement that reduces bandwidth and computational overhead at the pool-miner interface translates to measurable cost savings at scale.
The governance implications of Stratum V2 adoption are significant against the backdrop of increasing regulatory attention to mining operations.
In May 2021, Marathon Digital Holdings mined a "clean" block that adhered to U.S. anti-money laundering (AML) and Office of Foreign Assets Control (OFAC) standards — excluding transactions from sanctioned addresses. The incident prompted sustained debate about whether mining pools operating in OFAC-compliant jurisdictions could be compelled to filter transactions at the block template level.
Under Stratum V1, such filtering is technically straightforward: the pool operator constructs the template, and compliance requires modifying a single software instance. Under Stratum V2 with Job Declaration enabled, filtering would require either compelling thousands of individual miners to modify their node configurations or rejecting miner-constructed templates that include flagged transactions — a far more visible and contentious action.
Matt Corallo, writing in a Reddit AMA, framed the significance: "This is huge for mining centralization. Instead of being focused on the centralization of pools (which is the world we're in today), we can focus on the centralization of actual miners [and] farm owners!"
The distinction matters. Even if hashrate remains concentrated among a few large mining companies, distributing template construction across those companies' individual machines and operators creates a different censorship-resistance profile than concentrating it at the pool level.
However, this is not a complete solution. If a pool rejects miner-constructed templates that include certain transactions, the miner must either comply, switch pools, or solo mine. The economic pressure to remain in large pools for consistent payouts remains. Stratum V2 shifts the censorship question from "will the pool filter?" to "will the pool reject miner templates?" — a politically more costly action, but not an impossible one.
What changes:
What doesn't change:
The CoinDesk analysis noted this explicitly: "Stratum V2 does not change hashrate concentration, but it does change who decides what goes into each block."
The Stratum V2 Working Group expansion represents a structural commitment by the dominant mining pools to address the protocol layer that has concentrated block construction power since 2012. The economic case — efficiency gains in a margin-compressed post-halving environment — may prove more decisive than the governance case, though both point in the same direction.
The gap between Working Group membership and production deployment remains the central uncertainty. Braiins and DMND have demonstrated the protocol works in production. Whether Foundry, AntPool, and F2Pool will ship Stratum V2 support to their miners — and whether those miners will run their own nodes to take advantage of Job Declaration — will determine whether this is a structural shift or a symbolic one.
The protocol does not solve hashrate concentration. It does not eliminate the ability of pools to influence block content. What it does is make that influence visible and contestable, shifting the question from whether pools will filter transactions to whether they will openly reject miner-constructed templates. In a network where economic incentives and social norms have historically constrained bad actors more effectively than technical mechanisms, that shift may be sufficient.