Your suggestions are reasonable, and perhaps we can implement them in the next step.
The first step is to implement the GMX V2 framework.
The second step is to continue to implement the possibility of having both V1 and V2 methods of providing liquidity within the same GM Market. V1 is what you mentioned, where a single pool provides liquidity for multiple GM Markets while limiting OI.
We will further research this mechanism and look forward to discussing it with you further after forming a more comprehensive technical document.
I’m going to be uncharacteristically un-canadian and not be polite, and instead call bullshit on large parts of this post, other than saying yes an investment of $200k-400k is not insignificant and should be discussed and is being put to the DAO for a vote since it doesn’t cover development of the existing EVM GMX v2 protocol.
Would specifically love to double click on this sentence, i think its a baseless statement i’m trying to think of where GMX has footed the bill for other projects as you term it and some promise of huge dividends. My guess is you are grossly mischaracterizing the Partnership Swap program (Partnerships) as some sort of giveaways.
If your concern is with the Partnership program, or the way the contributors have managed that for the DAO (since you seem to be implying either poor use of funds or bad management?) happy to continue that discussion in a separate thread.
I
If you start 50+ markets with separate LP pools, it may be too late to migrate to the Universal pool because the first impression of trading and LPing matters.
There are a lot of good perp dexes. What makes GMSOL so special?
I can name: a good-looking UI, moderate trading and funding fees, and a share of GMX’s reputation and brand.
But is it enough? GMSOL does not have inflationary rewards, ecosystem grants, or a userbase like Jupiter.
It doesn’t have external funding, pocket marketmakers, or deep liquidity ready on day one.
GMSOL actually has to fight for survival.
Hyperliquid is the top DEX, not because it is a state-of-the-art project. It just had pair variety, liquidity, LP simplicity, and points. Hyperliquid was expensive (high funding fees and strict liquidations), but it succeeded anyway.
———————————————
II
I have no commercial interest in GMSOL’s success. I just have a warm feeling for GMX as the venue that has earned me a bit of money in the past.
So, being neutral, I want to brainstorm about possible GMSOL killer features.
You have a good-looking UI, moderate trading and funding fees, and a share of GMX’s reputation and brand.
But what about killer features?
It might be:
A wide choice of trading pairs with sufficient liquidity. More than Drift has, way more than Jupiter has. Price impact can be configured to mimic the CEX liquidity of each coin.
Fast listing, Hyperliquid-like. All coins that are listed on Binance, OKX, Bybit, and Coinbase are listed on GMSOL instantly; liquidity for trading is available instantly.
Easy LPing with 3-digit ARP (of at least 3-digit APY). It can only be achieved with high utilization.
Using GM LP as the DeFi LEGO brick, JLP-like.
Accepting yield-bearing USDe as the cross-collateral for trading, so users will trade and earn simultaneously.
————————————————-
III
Actually, I’ve just thought of an even bolder suggestion: to power ALL the GMSOL liquidity through a single GM pool backed by SOL.
It will drastically improve simplicity, APR, and DeFi LEGO usability.
GM LP will outperform almost any asset during the bull, while the risks can easily be curbed through OI limits and price impact.
BTW, all the SOL liquidity can be staked in a diversified basket of reputable LSTs, delivering solid additional yield.
As I am thinking about it, it looks more and more juicy for me as an old degen liquidity provider.
———————————————————
Anyway, this was just my thoughts.
I wish you luck with your project.
I fully support this proposal’s phase 1 (implementing V2 on Sol), but have questions/comments regarding next steps.
TLDR: I believe GMSOL should focus on long tail assets with custom pool behavior as described in X4/PvpAMM proposals to unleash next-gen memecoins on Solana. Leave leverage trading of deeply liquid bluechips on GMX and on orderbook-based private chains.
GMX has built a reputation of being robust and resistant to exploits and hacks in order to protect the 100s of millions in protocol liquidity, both for V1 and V2. However, this focus on security comes at increased time for testing and development. Therefore, releasing GMSOL as an affiliated protocol on Solana makes sense. It would allow faster iterative development in order to fully capture the opportunities of the bull market.
I think translating V2 contracts for SOL and testing them out is good as it is the building block of everything else. However, I’d like to question what do people on Solana actually want to trade? And what should the end-goal be for the GMSOL platform?
From recent Solana activity, I’d say that people would rather trade ‘long-tail assets’ than bluechips on leverage. In that sense, GMSOL should aim to be a testing grounds for GMX Swap and PvP AMM as they were described in the X4 substack proposal.
Also, trading on Solana is targeted towards retail traders, whereas Arb is targeted towards cautious whales that don’t want to send their capital on a CEX or a private-chain DEX such as Hyperliquid, Dydx or Apex. Therefore, I don’t think depth of liquidity will be a critical issue. Pools should fill sufficiently with real yield and points. Even more so if the focus in on long-tail rather than bluechips.
I’m also a proponent of buybacks rather than real yield at this point. Both mechanisms can coexist, similar to ETH’s deflationary model.
Concerning the comments on partnerships. Those are needed. That’s how you grow and cement your lead. With allies. In Game of Thrones, the Lannisters needed the support of the Tyrells and the other great Houses of Westeros to keep their place on the iron throne. Same goes to forging alliances with other projects within DeFi if you want GMX to stay on top. Partnerships go both ways.
Audit fees are significant but I think it’s natural that GMX’s treasury provides support for this. Perhaps could be structured as part loan, part grant if some members are very worried about this.
Great work and it seems that you are very invested in this project, Q. Will be following closely!
I feel very fortunate to receive this quality of advice and perspective before the launch of GMSOL. You will take GMSOL to the next level. I have submitted this to the technical team for internal discussion, and look forward to continuing communication with you.
Thank you very much for your detailed analysis and comments. Your insights into Solana’s user profile are profound. Indeed, we hope to provide a sufficient number of trading markets to explore more possibilities and boundaries. Only through full utilization can we identify directions that require further implementation. Solana, besides having excellent network performance, indeed has one of the most radical user groups in DeFi, with a huge demand for emerging assets and the highest risk tolerance. I believe such a deployment should also be able to fully test GMX’s carrying capacity through GMSOL.
Therefore, providing liquidity and ensuring a good trading experience for long-tail assets indeed require deep consideration and complete implementation. Fortunately, in this regard, we are very able to receive many constructive opinions from many members of the generous and wise GMX community, which need to be gradually discussed, tested, and evaluated for feasibility.
As for buyback and real yield, yes, I agree with you that they can coexist. Since GMX already has real yield, we can fully implement a high proportion of buyback for GMX on GMSOL, while observing the impact of buyback on GMX and minimizing the impact on GMX’s own structure. Focusing on incremental value is also the basic theoretical support for my conception and proposal of ARB incentives, SS Pool, GMSOL, and a series of upcoming proposals.
As for partners, I also strongly agree with your point of view. I also like Game of Thrones. For GMSOL, there may not be competitors. GMSOL is just a brand-new idea, and every mature protocol is a mentor to GMSOL. In fact, GMSOL will be open-sourced before its official launch. There may be a GMX-Solana-Core crate, and any protocol can access the GMX-Solana-Core interface on its own. We can provide a standard implementation of the GMX model in Solana together, or further, a general framework for the Trader-LP model. For different protocols, the final difference lies in choice rather than implementation.
We are willing to listen and learn. We believe that under the brainstorming of community members, there will be sufficiently good ideas. We also believe in our own innovation and implementation capabilities. We believe that open source and decentralization are necessary choices for a better future.
When you say things like “may one day hold millions of GMX tokens” it makes all this seem less legitimate. I understand being optimistic, but such a claim would mean generating hundreds of millions in fees (in order to acquire the GMX tokens on the open market, with 60% of fees going to buybacks). That indeed would be great for GMX, but with zero track record, it amounts to just a very, very optimistic “what if”.
GMX is currently about $30. There are less than 2M tokens in circulation.
Yes, was referring in large part to the partnership swap program, which lacked oversight, auditing, and most importantly lacked any material returns for GMX. Or in your words, poor use of funds and bad management.
All of those proposals were based on promises, none of which really materialized.
When GMsol promises perhaps acquiring “millions of (unstaked) GMX tokens”, I use the past partnerships as reference, where the DAO was in effect just gifting yield bearing tokens to subsidize the treasuries of other projects.
If GMsol is ready to fund their own audit, and GMX still acquires all of the benefits of their buyback program and the GMX DAO controls the treasury regardless, then the lowest risk option for GMX is to let the project proceed on its own. Mayeb I am misunderstanding the proposal, and these benefits for GMX are dependent on the DAO funding the audit. But if this is the case, then a more thorough accounting of possible fees generated (and thus buybacks) is warranted before moving forward.
Please note the usage of the words “promise” and “perhaps”. The expression “promise perhaps” does not exist.
You mentioned the lack of track record, which is natural. How can we prove the existence of something that does not yet exist? In terms of the fees generated, we don’t need to be pessimistic or optimistic. In my view, under the same model, it is basically impossible for GMSOL to exceed the fees generated by GMX, which is around 300 million USD in about 2 years. In other words, the fees generated by GMSOL will be greater than 0 but less than 300 million USD.
Another question is that you may still focus on what the GMX DAO will actually receive, the deterministic yield. I respect your opinion, but in my view, most GMX DAO members are not concerned about this short-term, visible, deterministic yield, but rather the long-term growth potential. Here’s a way to help you envision this “impact”: in the absence of GMSOL, if the current real yield of GMX is removed and entirely changed to buyback, what will GMX gain?
The reason for drawing the above conclusion can be seen from the recent two votes on MP. A large number of MP holders, even whales with a 200% boost, voted to terminate their own MP benefits. The resolution to End MP Rewards was ultimately passed with an overwhelming advantage. This is the spirit and vision of the GMX community.
If the focus was on short-term interests, the vote would certainly be for No Change. The result would not be 90%:10% to End MP Rewards, but 90%:10% to support No Change. If it were the latter, I believe there would be little hope, and I would be too lazy to devote more time, energy, and resources to contribute to such a community. But the GMX community is not like that. The GMX community is the one I consider closest to the early Bitcoin community, and X is one of the people I consider closest to Satoshi.
This is also why I am very proud to be a member of the GMX community and willing to propose such a generous offer, even paying a large amount of upfront costs to promote the initiation and development of GMSOL. Perhaps more details don’t need to be disclosed. Anyone with a little experience will know the difficulties and various investments required here.
The audit fee indeed needs to be approved by community vote, which is the purpose of this proposal. I want to say that 200k-400k USD is indeed just an insignificant number for me personally, but I am still very touched that X can propose letting me make a proposal for the GMX community to vote on whether to support the DAO paying for this equally generous offer. For me, this is an affirmation, and its significance far well exceeds the audit fee itself.
Without X’s vision and dedication, there would not be so many like-minded friends in the GMX community. As mentioned earlier, GMSOL could choose to raise funds but did not, GMSOL could issue new tokens but did not. All the value is given back to GMX. I hope you can understand this trade-off and vision more deeply.
This is a vote. You can choose to oppose or support it. This is your freedom. In any case, I respect your choice.
Overall, I think the high-level has promise.
My concern is about the actual value accrual mechanism for GMX holders.
If GMSOL Points (which aren’t owned by GMX DAO) hold claim to the GMX in GMSOL’s treasury, then it’s not really a buyback and burn. GMSOL Points holders just got the fees, they just swapped the native assets for GMX. So GMX holders get some micro-benefit of a whale (GMSOL treasury) market buying GMX, but none of the long-term benefit of a buy-and-burn (which permanently deflates supply). Promising not to stake the GMX helps a tad, but that can change.
What the GMX DAO is giving here is not just the $200k-400k for the audit (which I agree is insignificant in the long-run for a solid opportunity to beachhead onto Solana). The more valuable asset is a license for the V2 tech. The GMX DAO needs to ensure that it receives fair value for that license, as the tech took a lot of time and $$$ to develop, and because GMSOL will be an indirect competitor, and may cannibalize a small % of volume.
In that light, I think that the audit cost should be covered by the GMX DAO, as the price for supporting this expansion, but ONLY IF the upside will substantially benefit GMX holders in the long run. This should be framed as a high-risk, high-reward bet from the community with asymmetric upside.
I understand that the GMSOL team needs to eat. Both to pay for ongoing development, and to have profit motive. In that vein, I’d like to see clear mechanics as to how GMX holders benefit via buyback-and-burn, or real-yield. Both are very direct in terms of long-term benefit. Simply market buying to push up prices won’t help in the long term if those bought tokens are just being accumulated to a non DAO-owned entity that could later stake them or re-sell them back into the market.
All that said, I recognize the significant effort required to port GMX to Solana. I’m in favor of GMX getting onto Solana. I know that it won’t happen via the core team due to competing priorities. So my position is generally favorable, but contingent upon the arrangement being fair and mutually beneficial in the long run. GMX owns the tech. Let’s ensure that GMX gets paid for it, in a way that still gives significant benefit to the GMSOL team.
There is one point that needs to be clarified. GMSOL will not directly control the GMSOL Treasury. We do not intend to nor seek to control the GMSOL Treasury. In fact, I have proposed to the GMX Contributors that a multi-signature management mechanism should be established to ensure the safe operation of the GMSOL Treasury.
The purpose of the GMSOL Treasury may change. As the development stage evolves, it may be given more functions, but the change in purpose is focused on how to better improve the efficiency of capital utilization and how to use this reserve to promote the development of GMSOL, rather than profiting through any form of selling.
The initial function of the GMSOL Treasury is designated by GMSOL according to the development mechanism. For any subsequent necessary changes, GMSOL will initiate a vote in the GMX DAO. If agreed upon, it will be executed; if not, it will remain unchanged. In other words, beyond the initial functions, GMSOL cannot arbitrarily change the usage mechanism of the GMSOL Treasury. The aforementioned multi-signature structure ensures this.
Moreover, GMSOL and GMX will not constitute direct competition. Firstly, the overlap between Solana and EVM users is very low. Secondly, for GMX, the highest priority and efficiency in the short to medium term may be horizontal expansion within the EVM ecosystem, which should be a clear consensus.
As for how GMX benefits from GMSOL, I hope everyone will look at this issue dynamically from two perspectives: First, as long as GMSOL Treasury > 0, the reduction in GMX’s circulating supply is a positive gain. You cannot and should not view the issue statically from the end of history, or everything would be meaningless. Second, GMSOL has clearly stated that it will not have its own token. If GMSOL does really well in the future and gains market attention, if you want to share in the benefits of GMSOL’s development, if it is through buying tokens, your only choice is GMX.
Additionally, I am clearly opposed to the method of buyback and burn. Compared to deploying the same magnitude of assets passively in the GM pool through the GMSOL Treasury, which can simultaneously reduce circulation and promote initial adoption of small markets and many other functions, buyback and burn can only reduce circulation. This is a huge waste of capital.
Then, all our discussions are based on the hope that GMSOL’s operation will be built on a foundation of fairness and transparency as much as possible, avoiding direct conflicts of interest with any user group, whether it’s GMX itself, GMSOL traders, or GMSOL LPs. If you can think of any way that GMSOL has the opportunity to act maliciously or do things that harm users, please discuss it in the GMSOL group. If there is a solution, then our highest priority is to implement that solution. If there is no solution, then our highest priority is to come up with that solution.
I respect your prudent consideration for protecting the value of GMX. If there are any other questions, I look forward to further communication with you.
Overall, mentioning X many times may be because I am trying to answer every question raised by community members, especially the questions from cryptoptera that repeatedly involved the audit fees. Each time, I expressed my same gratitude to X. If X can express something, I think it would be good, but I am not sure if it is convenient for him to express directly. In any case, I also respect this, because GMSOL is ultimately an independent project.
This will be a proposal and vote initiated by a GMX_DAO Delegate. Whether it is supported or opposed, we respect the rules of GMX_DAO. Direct statements from Core Contributors are welcome but not necessary, as this is not their decision.
I got to say, your commitment to GMX is uncanny. Its like getting long form xdev passages, when x wrote more in the early days.
What I respect and get most excited about is that the work is ready for audit and that you are willing to pay if the DAO rejects paying for the audit fees, GMSOL treasury gets managed conveniently by GMX DAO/Contributors, the fees buyback GMX, and (correct me if i am wrong) distribute that GMX to LPs and traders from the points system.
Not a solana user… until GMSOL
PS: apologies as i have nothing critical to note, and thanks to those who have.
PSS: Q’s overall demeanor is restorative for the gmx community even before GMSOL.
I am deeply touched and grateful for your affirmation, which means a lot to me. These are not easy times, and we have encountered many challenges. The simplest way would be to avoid them or wait for others to resolve them; however, we are more willing to choose to truly contribute our own efforts, do what we can, and attempt to solve the problems.
After publicly releasing this proposal, as you have seen, there are a great number of very lengthy comments, both affirmative and skeptical, with very sharp questions as well as highly constructive suggestions. I am also striving to be a good customer service officer, answering everyone’s questions in as much detail as possible. In the process of responding, I can also better re-examine the perspectives of GMSOL and even GMX, reflecting on where there is still room for improvement, which has been extremely rewarding.
Since GMSOL was proposed, we have been working every day without weekends or holidays. As we further implement and research GMSOL, we can see even more possibilities, which makes us very excited. At present, we have compiled some very critical and innovative requirements, and next, we will enter a more intensive closed development period.
We look forward to the brand-new GMSOL meeting everyone as soon as possible!
one small thing related to SOLANA network congestion last (Apr?) , any UX effort planned? I do remember Jupiter did perform much better than others dapp during that period.
As to one of jaoda’s reply, that stable coin like USDe.USDT.USDC usage/design is essential to attract capital during startup.
Thanks for sharing again. Best luck to GMSOL which it does need since perp market getting crowd enough already. Also would be glad to see GMSUI etc soon (heard simialr lauguage base? I might be wrong)
I’m sure a lot of work went into the engineering here, good job on that.
Based on what I’ve seen on the market since GMX v2’s launch, the design of GMX v2 does not seem well suited to list long tail assets. I believe this lack of asset selection has held GMX v2 back from scaling. I also think the friction around LPing into a single market has made it harder to attract TVL for these long tail markets.
While I think its a great idea to see GMX expand to Solana, I would hope some more thought go into desigining a better mechanism that would allow for:
More asset selection for traders
Easier LP experience
More margining options for traders
I am concerned that by replicating the existing GMX v2 design on Solana that you will be far behind the other perp offerings already available on Solana. And as such it may be a wasted effort.
The recent Solana upgrade respects the priority fee, and the current solution is to provide additional Compute Units or declare a higher CU price based on the actual situation.
It might be possible, but surely we will focus on GMSOL for now.
First, I believe your concerns are very valid. We have already almost implemented GMX V2 on Solana, and it could be ready for audit if we were only aiming for GMX V2.
However, we share some of your concerns. Therefore, we are making efforts to achieve more asset selection and an easier LP experience, which will give GMSOL a mechanical advantage in attracting liquidity, especially for long-tail assets.
So, as you mentioned, directly replicating the GMX V2 design on Solana is far from sufficient. We fully agree with the three directions you pointed out, although the third point is something we are considering for the first time. We would like to ask if you have any specific suggestions regarding these three points, preferably based on the implementation of GMX V2. This way, the modifications to our existing framework would be smaller and less difficult.
(heard simialr lauguage base? I might be wrong)