What is Aptos AIP-146? Is it good for APT?

AIP-146: Please-driven high gas transaction limits

AIP-146 allows any user who controls a sufficient pledge amount of $APT to push transactions at a limit up to 100 times the network's normal gas limit. Overall, this is good for tokens: it provides major players in the DeFi space with a real reason to pledge APTs, rather than just gain revenue. The proposal is not an out-of-reach roadmap project-it has been accepted, an on-chain governance vote to enable the feature has been approved by a near unanimous vote, and Aptos Labs (@AptosLabs) has begun promoting the upgrade this week.

What is AIP-146?

Currently, every Aptos transaction is run under the same execution and I/O gas limits. This is fine for exchanges and transfers, but limits high-load operations: DEX clearing, risk engine updates, large-scale state migrations, and complex financial logic that requires more computing resources in a single transaction. Teams that reach the upper limit will either split their work into multiple parts or move parts of their processes down the chain.

AIP-146, fully known as "Pledge Based Trading Limits," written by George Mitenkov (@GeorgeMitenkov) of Aptos Labs, provides an optional fix. Transactions can carry information requesting a higher limit, expressed in basis points, ranging from slightly above 1 to 100 times the upper limit. To qualify, fee payers must demonstrate control of a large number of committed APT pledges. In this first release, only execution and I/O limits can be increased, and storage costs remain the same.

This privilege comes with built-in costs. Any high-limit transaction based on pledge must pay at least 10 times the minimum gas unit price, which deliberately sets a brake against spam and abuse. In addition, high-limit transactions will not receive special treatment in the memory pool.

How does the pledge level work?

There are three ways to prove a pledge: owning a pledge pool, becoming a proxy voter in the pledge pool, or holding a pledge in the pledge pool. Pledge of commitments includes active balances and inactive balances to be processed, and is verified based on the real-time status on the chain.

Genesis configuration has three levels, with independent thresholds for execution and I/O:

2 times multiplier: 1 million APT (execution) or 5 million APT (I/O)
4 times multiplier: 5 million APT (execution) or 10 million APT (I/O)
8 times multiplier: 10 million APT (execution) or 20 million APT (I/O)

The I/O threshold is higher because state reads and writes are more expensive for the network than pure computing. Request to round up: If you require 3 times, you need to reach 4 times the pledge amount. Governance agencies can add, remove, or adjust levels at any time without changing the code, so the 100x cap is reserved space for the future, rather than a feature that is available from the first day.

Will AIP-146 lock in more APTs?

This is what makes the proposal interesting for APT holders. For any protocol running serious on-chain workloads, access to higher quotas is valuable, and the only way to do so is by committing pledges. This has prompted DEXs, perpetual contract platforms, market makers and institutions to tend to pledge or arrange pledges in addition to their usual rewards and governance rights.

Aptos has been redesigning its token economy model in recent months: a hard cap on total supply of 2.1 billion APTs, current supply of approximately 1.2 billion, transaction fees destroyed, and governance to reduce the basic pledge reward rate to approximately 2.6%. Linking network capabilities to pledges provides large holders with new motivation to lock in APTs rather than maintain liquidity while the attractiveness of earnings alone is weakening. About 770 million APTs have been pledged, accounting for about 64% of supply, so the mechanism is based on an already high benchmark.

Deployment is progressing rapidly. The proposal was written in April 2026, accepted in the AIP Warehouse in early July, and implemented through the On-Chain Proposal 200 "Enable Transaction Limits". The vote was passed with 369.9 million APTs in favor and only 1449 APTs against, with a voting rate of 30.7%. It was subsequently implemented on the main network. Actual adoption will still lag: SDK and CLI support need to be tracked separately, and high-limit transactions can become routine only after new transaction formats are added to the wallet.

Is it good for APT?

Overall it is good. This upgrade removes the technical ceiling for the applications Aptos attracts through its on-chain market push and achieves it in a way that transforms network capabilities into pledge demand. Large pledgers bear the actual economic weight and also have relevant interests, which is the logic of linking privileges to pledges.

The entry threshold is 1 million APTs, so this is a feature for agreements, funds and giant whales, not retail investors. It is designed to concentrate network privileges in the hands of the largest holders. The mechanism was rolled out as an experimental feature, and levels and access rights could be changed at any time through governance, so builders were told not to rely on guaranteed access rights. Still, the direction is clear: Aptos has just provided its most serious users with a tangible reason to lock in APTs, and pledge rates will show whether they will accept it.