Aptos’ validator footprint has narrowed, and the change goes beyond a simple drop in node count. A report carried by PANews said the shift reflects a combination of performance upgrades, reward design changes and a steep decline in the token price, all of which have pushed operators to rethink where they run infrastructure.
In October 2024, Aptos had 146 validator nodes spread across 22 countries and 48 cities. The network still had a broad global footprint at that point. Outside Europe and the Americas, nodes were present in Singapore, Tokyo, Seoul and Hong Kong, while Sao Paulo, Johannesburg and Sydney were also part of the map.
By September 2026, the number of nodes had fallen to 84, a 42% decline. Coverage dropped to 13 countries and 28 cities, largely in step with the fall in validator count. Most of the remaining nodes were concentrated in Europe and the Americas. Outside those regions, only Tokyo still had a node, and Aptos’ earlier multi-city presence in Asia had largely disappeared.
Faster chain performance raised the importance of location
The report said two things happened at once. Chain performance improved, making it harder for validators to operate reliably from distant locations. At the same time, lower rewards and a falling token price put pressure on validator finances.
The Baby Raptr upgrade in June 2025 and AIP-131, known as Velociraptr, cut Aptos block times to below 50 milliseconds. That was a clear improvement for users and transactions. For validators, though, network conditions and data center location were no longer secondary considerations.
Aptos calculates rewards based on stake amount, reward rate and validator proposal success rate. The farther a node is from the main validator cluster, the easier it becomes for proposal success rates to slip under the same staking level, reducing revenue. Nodes spread across different continents were therefore among the first to feel the effect of the rule set. Shutting down or relocating to Europe or the Americas became the more practical option.
Higher performance also raised hardware requirements. The report added that AI-driven demand had pushed up memory prices, increasing fixed costs for validators. Operators running at long distance, on a smaller scale, or with thinner margins were the first to struggle.
That points to a broader tension in the sector. Faster consensus can end up compressing geographic diversity. Latency is a physical constraint, and when rewards are tied to proposal success, nodes naturally cluster in lower-latency regions. In that setup, decentralization in geographic terms can shrink as performance targets rise.
Lower rewards and a weaker APT price hit validator economics
The direct reason for validator exits was still revenue. Validators earn in tokens, but they pay for data centers, bandwidth, operations and staff in U.S. dollars. Those cost and revenue lines move in different market cycles.
Aptos annualized staking rewards fell from 7% in October 2024 to 2.6% in September 2026, a 63% drop. The path was laid out in stages. AIP-119, proposed in April 2025, lowered the rate from 7% to 5.19% starting in June that year. A tokenomics adjustment proposed by the foundation in February 2026 then reduced it again to 2.6%. Validator income comes from commission on delegated staking rewards, so lower reward rates reduced income for both delegators and operators.
Price action put even more pressure on the model. Over the same period, APT fell from $9.50 to $0.58, a 94% decline. After the validator count dropped, average stake per node rose from 5.75 million APT to 8.97 million APT, up 56%. Staking became more concentrated, but the combination of lower rewards and a lower token price still cut annual rewards measured in dollars by 96%.
The report framed the math plainly: a 56% increase in stake could not offset a 96% collapse in revenue. That, it said, is the arithmetic behind the validator exit wave.
From a long-term dilution perspective, a shift toward lower issuance and lower staking rewards may have its own logic. The trade-off is that validator operations become thinner, and network distribution narrows with them. Tighter issuance on one side and heavier operating pressure on the other are difficult to balance at the same time.
Other proof-of-stake chains face similar pressure
The report said the mismatch between token-denominated revenue and fiat-denominated costs is not unique to Aptos. When markets weaken, operators in peripheral regions and smaller firms are often the first to leave. What remains tends to be exchanges, institutions and professional validators already based in European and U.S. data center hubs.
Ethereum has seen a related debate. EIP-8363 proposes burning part of newly issued rewards as staking scale expands, with the aim of curbing inflation and excessive staking growth. The direction is similar to Aptos’ reward cuts, though the report said validator economics would still need to be tested.
By that reading, decentralization cannot be judged only by node count and staking ratio. The more important question is whether a sufficiently diverse group of operators can still remain when token prices fall and reward policy changes. Node counts can expand in a bull market, and they can also be reordered by cost during a downturn.
At this stage, participants that can absorb short-term reward volatility matter more. Exchanges and institutions have support from other business lines and longer-term service demand, giving them more capacity to stay through a down cycle. That gives exchange staking and institution-grade validators a positive role in network stability. But a rising share for those players can also weaken geographic and operator diversity. Stability and dispersion do not always move in the same direction.
Another key issue is the barrier to entry. The report said most public chains focus engineering resources on performance and throughput, but they also need efficiency work: achieving similar performance with less hardware, lower power costs and lower bandwidth use. Lower barriers would give new operators room to enter and help existing ones survive under lower rewards. Otherwise, “global distribution” may remain a line in documentation while actual nodes continue to contract into a small number of cloud regions.
Three conditions will shape long-term decentralization
The report said Aptos’ changes over the past two years do not simply prove that decentralization has failed. Instead, they reveal a set of constraints: performance upgrades change who is suited to be a validator; rewards and token price determine who can still afford to keep machines running; and the operators that remain will shape the network map in the next phase.
- Performance upgrades change who is suited to be a validator.
- Rewards and token price determine who can still afford operating costs.
- The participants that remain will shape the network’s next map.
In the long run, the report said decentralization still depends on three things: validators that can keep running during weak markets, operating costs that can be pushed down effectively, and an entry environment that remains open to new participants. If any one of those is missing, node counts may still look acceptable for a time, but geographic distribution is likely to narrow first.