Buyback GMX: A Proposal for Reward Distribution Optimization

Being dependant on the price for adoption is the main issue I see with the proposal. We want an activity that isn’t dependant on the price action of GMX… We are already in the spotlight, I don’t think trying to artificially create buy pressure is beneficial in the long run.

Yep, I agree with you. “What you may gain in terms of small-scale consistent additional buying pressure, you may lose in ‘real yield’ narrative strength."

@natgmx while I don’t agree with your proposal I do want to thank you for taking the time to be an active member of the dao and seeing how to improve the protocol.

There is value in unifying the distribution asset of yield including across multiple chains and GMX could be an option for that, from a practical standpoint as @Saulius mentioned we have to remember we have over 50 protocols and integrations that would be affected by making such a fundamental change to the protocol contracts.

At a time when we should want this army of development resources focused on integrated and building on top of v2, instead attention would be on fixing a set of broken claim contracts, since auto-compounds into GLP no longer work since their rewards are in tokens that can’t be put back into the system, changes mean re-auditing their contracts and GMX appears a less reliable partner.

One small additional point not yet raised, as we have no people join the GMX ecosystem, especially for small depositors there is a great value in the claim process being in the native token of the chain which is that they are constantly accumulating smaller sums that is gas available for future claims, compounding etc…

Saying all of the above, I do think there is merit to seek in the v2 front end an OPTION to compound into GMX or GM tokens if that can be done in a simple fashion by the development contributors


@coinflipcanda Thank you for your detailed explanation. There are indeed many aspects that I hadn’t considered, such as the impact of such changes on the stability of the protocol, the need for re-auditing, and so on. It’s also true that having the native token as gas fees is crucial for small depositors. Based on this, I completely agree with your view that seeking an optional method in the v2 frontend to compound rewards into GMX or GM tokens has value, as long as it doesn’t affect the development of essential features in v2.

The core logic is that while replacing ETH with GMX has its advantages, it inevitably sacrifices some of ETH’s native benefits. The question my proposal raises, which I also wanted to know, is what are the native advantages of using ETH as rewards? In my opinion, many of the answers from users, such as ETH being more valuable than GMX and ETH being the universal currency of the internet, are biased and lack rationale. Your explanation has addressed the potential doubts in my proposal very well.

Overall, thank you once again for your response, which has dispelled my concerns and significantly boosted my confidence in holding GMX for the long term. I am also now able to withdraw my proposal and am proud that GMX has core members like you.


GMX’s ‘friendly fork’ on Polygon has coded the ability to compound earned rewards to their version of GMX and GLP. It may be relatively straightforward to fork that part of their code to GMX - something to perhaps consider.

This is not an improvement, it’s a drawback on UX. Many people love to accure passively Eth-Avax rewards through GMX. Adding one another step won’t help the tokenomics since nothing actually will change and it would damage the UX. Also some protocols/contracts might front-run GMX buybacks easily if we add this via smart contract.

However, i do like Metavault’s (GMX fork on polygon) claim function. They have compound option which u can chose to claim rewards as their token and directly stake it. This kind of improvement is a gud addition. But ur suggestion… i don’t think it is wise.


@Jonezee thanks Jonezee, I will check how it works.

@tervelix Front-running trades generally involve capturing large trading opportunities when you know in advance that sporadic and non-continuous buying pressure will push up prices. You enter the market ahead of time and sell your assets to the buyers. However, in practice, front-running such continuous and uncertain buying actions in a decentralized trading pool with high transaction fees would not be a good strategy, at least compared to holding. So, I’m not too concerned about the front-running issue, but if my understanding is incorrect, please feel free to correct me.

In fact, after coinflipcanda’s response, I no longer agree with my proposal either. His reasoning is quite detailed and verifiable. The main purpose of proposing is to provide the community with another perspective, and both positive and negative feedback can be beneficial. A well-reasoned discussion can deepen everyone’s understanding. As for the feature on Polygon you mentioned, it is very useful information, and I will look into it immediately. Thank you for sharing.


Thank you for your proposal and the reply here. I love this community everyone trying to add something to the protcol. We all want best of the protocol as and help as much as we can do. Cheers!

1 Like

I disagree. Purchase $GMX with ETH rewards, makes more sense.

I agree with this. Allow the user to make the decision to use ETH to be converted into GMX. I do that anyways but typically have to go to uniswap. What makes GMX unique is that we are paid in ETH. Everyone else pays in their own token. We don’t want to lose the real yield narrative.

1 Like

Overall I like the idea of having GMX run like a DRIP (dividend reinvestment). I would be curious for people in the US if using ETH rewards to then purchase GMX would be a taxable event.

I like the idea of being able to more easily convert accrued ETH rewards to GMX (it’s what I do anyway), but I believe GMX’s reputation is very good especially because we don’t do ponzinomics and inflation and rather it’s very clear the user is being paid in a “trust worthy” coin. Just add one more check to the compound modal, disabled by default, to convert ETH to GMX.


Agree with @Jonezee in not doing automatic buyback

Although there may be a middle ground, where x% of fees goes toward buy/burn.

Capital allocation should be a balance between (a) retention for growth, (b) dividends and (c) buybacks.

The launch of v2 is a great opportunity to begin rethinking our capital allocation priorities, which is why I am in favor of the desire for (a) in the “GMX v2: sustainable development” thread as well as adding the wrinkle of using any excess for (c) buybacks.

Add a reinvestment button where users choose to automatically buy gmx and stake


could minimum change to

1\stake MPs
2\claim GMX Rewards
3\stake GMX Rewards
4\claim esGMX Rewards
5\stake esGMX Rewards
6\claim WETH Rewards
7\convert WETH to ETH
add-8\convert ETH to GMX
add-9\stake WETH Rewards in GMX



very strong no from me:

I think what made the GMX model successful compared to earlier defi models (and made others follow) is that it resembles public market corporate structure. GLP operates like the bond: user loans the platform universally accepted collateral so they can perform everyday operations and receive a higher yield as a result. GMX operates like the equity: user receives less yield (though quite large compared to TradFi dividends) and is exposed to the volatility of the organization itself compared to other organizations (gains, mux etc) in how competitive it is in fees, liquidity, security, etc. and can be traded to realize gains in valuation, but similar to the ex-dividend requirement must stake to receive the yield.

  1. Changing the ‘dividend’ to be paid in the organization’s own equity can create potential problems where the stakooooor (and market participatoooors as an aggregate) may find themselves in a situation where in a market downturn they receive less and less eth/usdc for their GMX (at a time precisely when they would want/need ETH or USD to cover leveraged positions) and where the ‘dividend’ being paid ends up acting something more like a death-spiral convertible bond (see Luna).

  2. This would not necessarily create more buying pressure as it would lower yields to users who would have to pay additional fees to swap from GMX into a desired universal collateral (and an additional internalized cost to convert from ETH to GMX initially in the pool). These lower realized yields could lower GMX’s competitiveness amongst other perps, which in turn has the potential to destabilize liquidity and the overall platform.

  3. GMX is an ecosystem at this point. Such a drastic change to tokenomics would affect many other teams and projects.

I like the creativity, but strong no.


Same idea as me, don’t know when it will be online

This would be great


all that’s needed. Default is ETH.