mDAO Proposal: Voting on validators stake using on-chain gauges through the MNDE token

Moved the proposal to a newly created category Governance.


Now I might be missing an economic component of this but, would it make sense at all to offer validators looking to get MNDE in order to direct staking the ability to buy MNDE from the treasury increasing the amount of SOL held and staked by Marinade while also giving them MNDE to control the stake? Definitely increases inflation on the token price which I know is not many peoples favorite thing but would offer a compounding effect on the SOL delegation these validators are receiving.

I feel that the Token Exchange program that was announced recently partially solves the matter you’re bringing up.

It is a reality that MNDE is hard to acquire on markets due to slippage and the Token Exchange will offer another way to get in the game. If this Token Exchange proves to be a very good way to distribute MNDE, all holders will be free to vote for another round of a similar nature, maybe even directly addressed to validators.

To come back on the most brought up topic of this discussion, I also believe that MNDE acquired through work and contribution should have the same power and rights as any MNDE on the market. The decentralization objective must be guaranteed by the broader rules of the Marinade protocol: whether it’s a team member or a highly funded individual, both should never have the ability to skew the goal of Marinade as its rules are clear and unmovable.

I feel that 1.5% of the total stake is a reasonable limit to begin with, as with 10% of the stake controlled by those gauges, someone would have to acquire 15% of the total MNDE in gauges to max out a validator before running into this ceiling.

This situation gives a “maximum” that can be reached by any validator or actor, which I believe is healthy. Because the financial incentive to accumulate more MNDE will drop after this point, there will be a reduced incentive to accumulate more MNDE. @dobby I’d be really interested to hear your thoughts on this “high limit” as you advocate for a higher limit (or a separate one) for the gauges. Don’t you think that if someone owns 15% of the gauges and maxes out a validator, the financial incentive should start dropping for this actor in order to help with the decentralization of power and interests?
It’s also worth considering that an actor could still see value in buying MNDE beyond this point to simply take it out of the supply in gauges or direct stake to another validator that may offer financial compensation in exchange (bribes).

I resonate a lot with @brian_smith_0’s points regarding this, the focus should be on MNDE fueling healthy and performing validators in a setup that prevents stake concentration harming the decentralization objective. I think that if we all agree on broad rules for Marinade, we’ll pave the way to get there together.

Note: To answer’s @dobby question regarding how to create gauges, the process to do so should be permissionless. Meaning any validator/MNDE user will be free to create the gauge for an eligible validator. So any eligible validator will be able to have its gauge and use/market it :slight_smile:


Thanks for the response Cerba - Totally see how the exchange program solves this somewhat. I suppose the only thing is that swapping with someone like Raydium and holding RAY in the treasury doesn’t really push mSOL growth but does leave Marinade stuck with RAY while giving a decent amount of influence over to an AMM.

Raydium is obviously just an example, not sure if they proposal is focused on any community member or specifically validators? Need to go back and re-read that medium post.


I know that the details of the Token Exchange program are still being ironed out, but I believe there will a be a set of tokens that are accepted, and those tokens would have to be directly useful to Marinade and not only be a treasury diversification. Not all tokens can be considered for the exchange and I believe the focus will be on SOL/USDC as much as possible.

Nonetheless, even if RAY is not an acceptable token to be traded for MNDE in the context of the Token Exchange, Raydium or any other protocols could participate with another accepted token.

1 Like

Gotcha that makes sense - thanks for the clarification!

Hi everyone, I’m writing on behalf of a top 15 validator. I think this proposal would make Marinade uniquely different in the liquid staking landscape and build up a significant competitive advantage. It’s only a matter of time until stake delegation is financialized. Marinade should capture this opportunity, democratize it, and influence its ethos.

A few things that I think are true for the majority of top 20 validators:

  • Validators were the top contributors in many early Solana SAFTs and public auctions and thus own a significant amount of self-stake.
  • Many large stake accounts are owned by individuals that are part of validator companies or very close to them.
  • Having stake is our core business. Many products and services we built (e.g. MEV related) rely on the fact that we have stake. Decreasing that stake (e.g. converting to mSOL) runs counter to our core business interests.

On the other hand, siphoning off stake from the security group (top 33% of stake) is key to increasing Marinades TVL. Marinade would need to find a way to convince validators to convert at least some amount of self stake to mSOL. That isn’t easy, but I think it’s doable.

A key factor here would be the ability to recoup, to at least some degree, self-stake that is being converted to mSOL. Gauges controlled by MNDE is that ability. I was thinking about the 1,5% max stake cap and under which circumstances it could be increased as it will be quite a limiting factor in its original form. A possible way to tie the ability to influence stake delegation to the overall Marinade goals (increase mSOL TVL) would be to link the ability to gauge voting above a certain threshold to the amount of SOL converted mSOL.

E.g., a validator that wants more than 1,5% of max stake needs:

  1. Enough MNDE to influence the gauges. Building demand side for MNDE is crucial, I fully agree with the post of @Cogent_Crypto.
  2. An equal amount of mSOL in its wallet. Want 200k SOL more than the max cap? Gotta have ~200k mSOL. This is the tricky part - how can we make sure people can use mSOL in DeFi but still bind the ability to go beyond the 1,5% max stake to holding mSOL?

I think that way, we would be able to force an increase in Marinades TVL, create a strong demand side for MNDE and allow validators with a large self stake to get behind Mariande sustainably. It’s a classic win-win. Validators can convert self stake to mSOL without significantly decreasing stake on their validator. Marinade can sharply increase TVL and MNDE demand side.

How do we feel about that direction in general?


Thanks for this post.

From the 1.5% max stake perspective, this is a helper, not the end goal. I believe the market is smart in the long term, so perhaps instead of arbitrary conditions, 1.5% can be raised like 0.1-0.3% each month automatically. This should create simple and approachable conditions for everyone while still setting up an advantage for decentralizaton in the beginning.

Anyway, the max_stake parameter should be governable including removal. I would not advise to remove or significantly increase (10%+) it to the beginning when people will be still learning the process and consequences.


Thanks for the perspective and welcome to the forum!

As a note, I would say that how the various stake assignment limits could change is independent from the on-chain gauges proposal: on-chain gauges could fail, and a limit change proposal still pass; or the other way around.

I would personally recommend keeping proposals atomic, so perhaps forking this into a separate discussion topic might be best.

1 Like

Thanks for the replies @gekonn @PlayerOfBits - wholly agree with both of you.

Once gauges are live with cautious limits we can explore the real world dynamics and slowly work our way towards a desirable outcome.

We should consider marinatooor’s points carefully. I believe strongly that finding a way to incentivize large self stakers to use mSOL and defi would help further Solana’s decentralization more so than anything else. It’s very obvious the huge % of stake that is out there is from large self staked wallets when you look at these graphs from Solana Compass

That’s how we get to marinade TVL to be 10s of millions of SOL.


Sounds good! Do feel free to open a topic on the limit changes before gauges even goes to a vote, though, either here or on the Governance channel on Discord - I’d expect you to get a bunch of different opinions on that one, so it doesn’t hurt to have more discussion time.

1 Like

Let me bring a point of view that I think no one brought up yet:

  1. I think it is important that the contributors who receive part of the team allocation are able to keep their NFTs locked. We are building Marinade, we put lots of effort in it (and I think it shows) and we should be able to help steer the Marinade’s direction in the future as well - with help of the community and our partners - through the governance. All in all, being part of the allocation program means that we have even more skin in the game and are strongly incentivized to keep up the good work.

  2. However, receiving NFTs with locked MNDEs is considered income (at least where I live) and I am not sure I, personally, would be able to afford to keep all MNDEs locked as the taxes from this income could not be covered by a regular monthly compensation.

  3. I did not think about the team allocation this way before but now I realize that a way to utilize locked MNDE to ensure I do not have to sell is gonna be a necessity - e.g. by running a validator with stake delegated to it by the power of MNDE.


I agree with @Cogent_Crypto on this one. Focusing too much on encouraging validators to hold mSOL vs Solana DeFi users to use mSOL will lead to an eventual decrease of demand and the entire delegation being validator funds who could self stake instead.

1 Like

Thanks for coming back to answer this one! Really like the approach of making it permissionless for eligible validators to set up their own gauge. The only other thing I would say here is that there will likely need to be a strong system in place to cross-reference the list of eligible validators and ensure that it is kept up-to-date.

I do agree with you here @Cerba!

With respect to the 1.5% cap, I went back and did some more research, and I can now see where you’re coming from. If you look at the staking status updates that the team pushes to GitHub for each Epoch, there are only a handful of instances where the amount of algo-delegated SOL even comes close to the 1.5% max cap. (with the exception of Epoch 272 which appears to be a significant outlier).

I spot checked every Epoch published in the GitHub repo and only found ~ 8 Epochs where at least one validator was eligible to receive > 1.25% of Marinade’s staked SOL.

I will, however, still advocate for increasing the 1.5% cap over time as @gekonn suggested. I also think that 0.1% per month would be a great starting point. The more utility the governance token has, the more we all win over the long term.

Curious to see what others think of this idea too!


Hello, governoooors.
I think it’s time to suggest a final version of this proposal! This version will stay open to comments for 24-36 hours before being submitted to an on-chain vote!

What do you propose?

As a final proposal, Marinade would implement a voting system on validators stake using on-chain gauges through MNDE for 10% of the total Marinade stake (a governable parameter in the future).

The requirements for a validator to be eligible to a gauge would be:

  • The validator is not part of the superminority
  • The validator is not blacklisted for cheating with credits or changing its commission before the end of epochs
  • The validator has a commission equal or inferior to 10%.

As long as a validator respects this set of criteria, his rank (depending on the on-chain score calculated by the Stake-O-Matic) would not matter for the gauges.

This implementation will take into account the credits observed for the epoch (an on-chain metric) to guarantee the minimum performance of the validators getting stake through gauges. If a validator has less than 20% of the average credits observed for all validators, it would lead to an emergency unstake that overrules the stake obtained by the gauges. If the validator performs on the next epochs, it would get back the stake it should get from gauges.

Gauges would be a “Set & Forget” approach to stake delegation, meaning they would only require changes when a user gets additional MNDE or decide to change their allocation. The creation of those gauges would be permissionless and would not require a separate governance vote.

Gauges would not allow to go above the Marinade limit on stake attributed to a single validator, which is currently set at 1.5% of the total stake.

What is the rationale behind the proposal?

As explained in the original post, this proposal would bring another way to validators to attract stake as well as provide utility to MNDE. I invite everyone to read the original post and the discussion to get more context on the proposal.

What is the expected positive impact of this change?

We expect that those gauges allow validators to look for support from MNDE holders or acquire it themselves in order to secure stake their way in a sticky manner.

We also expect that this change will allow validators that are currently struggling to get into the top 300 in terms of score to unlock some stake from Marinade and start building towards a more performant validator.

We can also estimate that those gauges will allow MNDE holders to build or support new validators on the Solana blockchain, helping its decentralization and censorship-resistance overall.

Finally, this proposal has the additional impact of bringing validators to Marinade and make them part of the DAO, as they have financial incentives to hold MNDE.

Any other considerations?

The community brought up a lot of good ideas, which can be implemented independently of this proposal, and I would like for some of them to keep being discussed. I specifically think about Marinatoooor’s proposal on linking the ability to gauge voting above a certain threshold to the amount of SOL converted to mSOL, or on the proposal to limit gauges even further to smaller validators. I believe those ideas deserve to be discussed on their own forum thread and explored individually.


I am in favor of having this proposal voted on chain

1 Like

This is what it’s all about. Also in favor of moving this proposal to on-chain vote.


The proposal is ready for your on-chain voting pleasure.


Thread closed: Validator Gauges proposal passed through a on-chain vote, and is now live here