JIP-28: Accelerate BAM Adoption

[drnick]

JIP-28: Accelerate BAM Adoption

Category: Treasury

Abstract

JIP-28 proposes a targeted delegation program designed to accelerate adoption of the Block Assembly Marketplace (BAM) among validators by allocating a portion of JitoSOL stake to validators running BAM software.

This proposal introduces a structured, tiered delegation model that rewards early adopters, mitigates operational and economic risk, and ensures a gradual, safe transition toward network-wide BAM integration. This JIP extends the Directed Staking framework (JIP-27) by layering in BAM-specific eligibility and tiered delegation triggers that scale proportionally with BAM adoption metrics.

Motivation

BAM represents the next evolution of Solana block production — improving privacy, censorship resistance, and execution efficiency. Additionally, by increasing the performance of the Solana chain and introducing new fee streams from Plugins, BAM has the potential to significantly increase Jito DAO revenue – but only if the system is adopted by the majority of the network.

Validators who run BAM today take on operational risk to support the Jito Network’s long-term roadmap. The DAO should recognize this contribution and align its stake policy accordingly.

A JitoSOL stake delegation incentive allows the DAO to:

• Reward BAM validators for absorbing early-adopter risk.

• De-risk BAM onboarding by ensuring predictable rewards.

• Phase in BAM adoption safely.

Key Terms

BAM (Block Assembly Marketplace): A transaction assembly and execution system developed by Jito to enhance Solana’s efficiency, privacy, and fairness.

BAM Validator: A validator operating a BAM-enabled client that meets the performance and fee criteria established under JIP-27.

BAM Stakeweight: The percentage of total network stake running BAM, measured and averaged over two epochs.

Non-Directed Stake: Total value locked in the JitoSOL Stake Pool not subject to directed staking or JIP-28’s specialized sub-allocation. This TVL will be allocated across up to 400 validators that meet the criteria established in JIP-25.

Delegation Tier: Percentage of JitoSOL TVL allocated to BAM validators based on verified BAM network stake adoption milestones.

Specification

The passing of this JIP would lead to the implementation of a tiered delegation schedule that allocates an increasing amount of JitoSOL stake to BAM validators based on verified network adoption thresholds.The initial delegation would activate immediately upon passage of this JIP. Each consecutive tier activates once BAM stakeweight reaches the corresponding threshold for two consecutive epochs.

Delegation Criteria

Delegations will be distributed pro rata to all eligible BAM validators meeting the JIP-27 validator criteria (0% inflation fee, ≤10% Jito MEV commission, non-superminority).

Validators must run BAM continuously for at least three epochs before qualifying and maintain uptime and reliability metrics within 3% of chain maximums.

Once BAM reaches the threshold specified for 100% JitoSOL delegation, the BAM parameters shall be upstreamed to StakeNet for integration into the Non-Directed Stake delegation. This will eliminate the distinct BAM delegation category with permissionless delegation to any qualifying validator.

Governance and Implementation

BAM adoption percentages and validator eligibility are verified via StakeNet and Jito Foundation reporting. Delegations will be administered using the Directed Staking infrastructure as described in JIP-27. Operations will use epoch snapshots to determine eligible validators and stake distributions. Delegation parameters (tier thresholds, validator requirements, stake caps) are governable via future JIPs.

*Note: This JIP is contingent on the passing of JIP-27.

Temporary Pool Management Control

JIP-28 seeks temporary control of the necessary Steward parameters that will enable active and responsive pool management throughout the transitionary period. The same 6 month bounds on these powers and mid-term update will extend to this process. This will allow effective transition to the new pool economics, whilst minimising yield drag and inefficiencies in stake delegation throughout the roll out of these paired proposals. These temporary powers will be revoked at the end of a 6 month period unless a further requested via JIP and agreed by the DAO.

Benefits / Risks

Benefits:

• Accelerates BAM adoption: Direct stake support offsets early adopter opportunity costs.

• Strengthens Solana resilience: Encourages decentralization through phased onboarding and avoids sudden validator migrations.

• Reinforces Jito leadership: Aligns JitoSOL’s delegation strategy with Solana’s technical roadmap and Jito’s long-term mission.

Risks:

• Operational complexity: Manual delegation requires interim oversight before StakeNet automation. Mitigated by leveraging infrastructure created by JIP-27.

Outcomes

The passing of this JIP will lead to a structured, tiered delegation model that rewards early adopters, mitigates operational and economic risk, and ensures a gradual, safe transition toward network-wide BAM integration.

Cost Summary

This proposal incurs no direct cost to the DAO.

[Solus]

Why are we still forcing 0% commission when most of the network won’t be profitable running this, the minimum should be 5% with max as well, to allow nodes to be able to move their profitability where they need too. Stakers can move else where if it’s too high for them on the node.

Other pools have done this, if you want adoption and security in the pool/network. Raise the minimum, it’s ridiculous we’re even still talking about 0% commission.

[nick_figment]

Trying to understand how these tiers will work if JIP-27 is adopted (which introduces the 10mm SOL floor for the non-direct delegations). Could you please clarify and walk through a few concrete examples for delegations, including cases of JitoSOL TVL above 10mm and below — for all the tiers and hypothetical validator counts?

Is the BAM tier defined as a percentage of total JitoSOL TVL, or is it intended to apply only to the TVL above the 10mm permissionless floor each window?

[drnick]

Hi @nick_figment the relevant section is here:

Once BAM reaches the threshold specified for 100% JitoSOL delegation, the BAM parameters shall be upstreamed to StakeNet for integration into the Non-Directed Stake delegation. This will eliminate the distinct BAM delegation category with permissionless delegation to any qualifying validator

I think the request for some deeper exemplar modelling of how this will roll out and interaction with the permissionless delegation is a very reasonable ask. I will come back soon with some more details.

[DanCrypto]

I appreciate the intent behind Jito’s current direction: rewarding efficiency and alignment with MEV transparency. However, I believe that the 0% inflation fee requirement deserves a second look - as expressed before (JIP 25 - although late to the party). In practice it creates structural incentives that favor concentration rather than decentralization.

The race to zero on fees sounds appealing in theory, but it narrows the field to only those validators who can afford to operate at a loss for extended periods - eg Helius. This dynamic inevitably squeezes out smaller and independent operators, leaving the network increasingly reliant on a handful of large, well-capitalized entities.

We’re already observing this trend. In fact, the number of active validators has been gradually declining, and with the recent Solana Foundation delegated stake adjustments, this may accelerate. If the incentive structure continues to reward only those able to sustain 0% inflation fees, it will likely further concentrate stake distribution over time. Fewer validators means greater risk - both technically and in terms of governance capture.

Instead of a 0% threshold, I would propose a more balanced approach:

• Set a 5% inflation fee cap as the qualifying criterion,

• With penalties or reduced weighting for those exceeding it.

This still promotes efficiency and accountability, while ensuring the validator ecosystem remains diverse, sustainable, and resilient.

Ultimately, we should ask ourselves: Do we want to build a network that rewards short-term cost-cutting, or one that sustains long-term decentralization and operational diversity?

[drnick]

The following update to JIP-28 is proposed:

A faster acceleration towards JitoSOL delegation is suggested with the following schedule. The more rapid ramp reflects the confidence in the engineering delivery of BAM that network stability can be maintained with greater amounts of stake sooner.

Additionally, you may have noticed a recent Security Council transaction to the Steward with a Programme Unstake Cap raise from 20% to 30%. This was executed to manage a large churn in the pool due to an unexpected number of commission rugging validators following changes to SFDP. With both JIP-27 and JIP-28 requiring greater amounts of stake cap management and maintenance of the “floor” established in JIP-27 it is anticipated more transactions such as this will be required. Consequently, JIP-28 now seeks temporary control of steward parameters using the same structure established in JIP-27 to enable the on-going and responsive dialling of steward parameters throughout the transitionary phase of JIP-27 and JIP-28. This will avoid excessive use of the Security Council structure for low risk transactions.

The following paragraph will be added to the specification section of the proposal if delegates and token holders approve:

Temporary Pool Management Control

JIP-28 seeks temporary control of the necessary Steward parameters that will enable active and responsive pool management throughout the transitionary period. The same 6 month bounds on these powers and mid-term update will extend to this process. This will allow effective transition to the new pool economics, whilst minimising yield drag and inefficiencies in stake delegation throughout the roll out of these paired proposals. These temporary powers will be revoked at the end of a 6 month period unless a further requested via JIP and agreed by the DAO.

[ValidBlocks]

Overall the proposal looks good for BAM adoption but there are a few concerns.

Before expanding the current pool validators set from 200 to 400 there was a race to zero for a long time already. After the set expansion to 400 there are already more than 50 validators running on 5% commission, which is a breath of fresh air considering that a lot of validators have already shut down their nodes because were running on loss.

Now a new race to zero begins for the new BAM pool where 0% commission is a must (not an option like it used to be). This rises following question:

The 0% commission enforcement will still be active once the BAM transition completes?

For a while there will be two pools, so how the delegation mechanics work if a validator is eligible for delegation in both. Let’s have this example:

We are in the initial phase where 30% JitoSOL BAM Delegation (currently pool TVL is 14M, so that’s 4.2M for BAM delegation)

Validator A current target delegation is 0.25% of 14M, that’s aprox 35K

Pool is reduced by 4.2M (reserved for BAM delegation)

Validator A new target delegation is 0.25% of 9.8M, that’s 24.5K

Validator A is also eligible for BAM delegation. Currently there are 60 validators running BAM client with 0% commission.

Validator A BAM delegation target is 4.2M / 60 = 70K

Does it mean that Validator A receives stake from both pools, 24.5K + 70K from the example above?

What will be the churn percentage for the new BAM delegation strategy?

Cycle will continue to be every 10 epochs?

Is there going to be a max validators set size for BAM delegation?

[Simple]

If JIP-27 passes, the pool will always retain at least 10M SOL for the Non-Directed Stake.
Because of this, the calculation described above seems slightly different in practice.

For example, with a 14M SOL pool TVL:

• 10M SOL is locked to the Non-Directed side

• Only the remaining 4M SOL can be used for Directed Stake (User-Directed + DeFi-Directed + BAM Directed)

Thus the “30% BAM Directed” is not 30% of the entire pool, but 30% of the Directed portion after the 10M Non-Directed floor has been preserved.