dYdX V4 Launch Incentives Proposal

Hey, @carlbergman

My personal opinion it’s important to include projected costs of developing such program in the proposal.

Grant will be funded from DGP but community should know the costs when voting for this proposal.

And for god sake please finally clarify what happened with that $300K. Did @chaoslabs used that funds already or not. Do they have to do any work regarding this payment, will part of this money be used in the development of the rewards program, and so on?

Come on, we trust this company to develop a very important aspect of the protocol. There shouldn’t be any misunderstandings.

1 Like

it looks like good, agree with you

1 Like

I voted Against this proposal due to:

  1. Lots of clarity needed
  2. Vague purpose, prefer more agile approach such as launching and migrate to v4 first, then monitor and take action then based on revenue and active users metrics per epoch.
  3. 20M$USD is very big sum of money, its ~3/4 projected dxdy platform revenue for 2023, why do we have luxury to spend too much money for incentive program?

CMS here, have been long term supporters of dYdX and very excited for V4. I like the initiative and agree that the incentivization process is needed to kickstart the migration/flywheel, but I disagree with some of the implementations and would like to see these issues addressed before voting yes for on-chain vote (didnt vote in snapshot was still formulating thoughts). Someone please address @RealVovochka and @CipherLabs questions about costs and what happened to the 300k. These defi treasuries are not a blank check to ask for $X after the fact, some transparency please.

Incentivizing Desired Behavior
Deposits: No. This makes no sense. Just incentivize trading behavior, why would you incentivize a derivative of trading behavior.
Staking: No, stakers make money with fees anyways. Why should they make more money? Like this just doesnt make sense. They will stake cause the yield is good and the yield will be good because you are properly incentivizing trading volume. The one thing I can see is staking dydx help reduce trading fees more than just holding liquid dydx. Say staked dydx counts as 2 shares and liquid dydx counts as 1 shares and then you see how many shares you have and count trading fees that way.
Governance: No. This is the most ridiculous one, most users are not sophisticated enough to understand the nuances between the various proposals. Now they just gonna randomly click yes or no to get some token? The users dydx should be attracting are the people who just wanna speculate on the app and degening with leverage on their favorite token. That is best for long term growth of the protocol. You can run gini coefficients on top defi protocols, is there really a correlation with governance voting and protocol growth? TIL we should pay people to vote. Please remove this from the criteria for incentivization, it is very poorly thought out.
UI vs API: I would like to take a moment to address the differences between UI and API trading. Not all volume is created equal, uninformed retail flow is much preferred than sophisticated mm arbing against some cex (given that there is enough initial liquidity which I agree we should have some portion of early rewards specified for MMers). Say I stick a limit order in because I like BTC at 28k. Some hft arber prolly going to pick me off as it is a stale order and will instantly be in the profit. They would already do that regardless as they make money. I would argue that they should not get additional dYdX tokens for actions they would do without incentivization. In this essence, if you really want to get retail traders, I believe the best way is to incentive UI trading and specifically allocation a portion of the rewards to UI traders.
MM Rewards: This is a hot topic as V4 will not have them initially and have leaned to the rebate system below. It does take time to integrate a new system for MMers so they should get rewards for migrating and trying a new system. To make sure there is enough liquidity there needs to be some rewards for resting liquidity like v3 or injectives. Honestly the 2-3bp difference between making and taking fees isnt enough to make, especially in the shitter markets if I am getting taking rewards, so there will end up being alot of hidden liquidity with limit orders which is probably fine? Not sure right answer here just wanted to highlight a few extra variables but for initial launch, some of the markets will have thin visible liquidity.

Formula Transparency
Introduce unpredictability by incorporating randomness into the season lengths, raising the bar against potential manipulations.
Would lagging the rewards solve this issue a lot better? The further removed the rewards are from the end of the epoch the less MM gaming is going on here. Say 25% released each week for entire next epoch. That being said, on the extreme end if the fees are very off from the token emission rewards, the people gaming this system are simply providing an arbitrage service, say monthly rewards are 10m in emissions but fees are only 5m and theres a day let in the epoch. If you are a dYdX staker, wouldnt you wish for that gap to be bridged aggressively? Perhaps a better approach then is to trim the rewards if the delta is too big? From a first principle standpoint, not every month is the same in terms of volume, some months should have more rewards and others should have less. Can look at top X perp cex exchanges and do some small relative volume adjustments each epoch for the ones with more/less volume.

We understand the need to provide rewards for smaller markets, ensuring liquidity and competitive trader UX. Whether asset-specific percentages should be public or confidential is a question we are evaluating, leaning towards transparency. A compromise could be randomly selecting percentages from a predefined range.
I feel like this can be better done, the smaller markets should get baseline rewarded like @Jordi said so ample liquidity exists. There should be additional fine tuning rewarded based on volume and there are times where specific coins trade more than others, imagine into EIP 4844 with say ARB or OP. Perhaps looking at top 5 perp cex exchange volumes and summing volumes and averaging OI across the epoch to get a sense of what was traded the most. Then distribute rewards in some formulaic weighting with that basis in mind and some slight randomization in here. That was the baseline is more accurate and you are rewarding the pairs that traded the most.

Sybil Resistance
I feel like something with hedgies can be done here. If we only focus on trading rewards and just scale everything linearly, it wouldnt even be advantageous to sybil like the current dydx V3 program. Its actually better to just have everything in one account for volume deductions.

Additional Questions:

  • when will the program start, immediately upon V4 launch beta, full mainnet, few weeks after?

  • I read that Chaos Labs does not have enough voting power to submit a snapshot proposal. Its 10k dYdX, thats like 20k USD. Are incentives aligned enough where someone with less than 20k skin in the game should be in charge of a 20m incentive program? I am not saying those with the largest amount of dYdX will clearly do a better job, but how you gonna ask for 300k+ USD and not even have 10k dYdX tokens.


Interesting comments about perhaps offering more rewards for UI trading as opposed to bots. It’s clear there is a lot more thinking that needs to go into all of this. I’m still of the opinion it should all be done through the native rewards programme too. I don’t see why we can’t just increase the native rewards and tweak the formula.


Previous Grant Work

Chaos Labs is a long-term dYdX contributor with a history of successfully delivering grants. This collaboration led to an engagement by the DGP with a contract to develop a comprehensive suite of tooling for V3, of which the $300K USD in question was part. This was previously answered here and here by the DGP. The scale of this project was significant, prompting us to allocate considerable resources, hire a dedicated team, and realign our operational roadmap to ensure its success. However, the strategic decision to build V4 on Cosmos necessitated a pause on this engagement midway.

Understanding that much of the technical and engineering efforts underway would no longer be relevant with V4, DGP requested we put the project on pause while exploring alternative productive scopes. Despite the substantial internal investments, resources, and commitments made, Chaos Labs decided to forego additional work from the agreement. This decision was made notwithstanding our early and robust deliverables, such as the Risk Portal, Asset Listing Portal, and preliminary research on Permissionless Market Listings, all of which showcased our dedication and pace of execution. We share this not in pursuit of empathy but for the sake of transparency. We fully recognize the inherent risks associated with investing in rapidly evolving and innovative ecosystems and continue to be proud contributors to dYdX. To be clear, the $300K USD expense from the V3 engagement is unrelated to our current or future v4 proposals.

However, it’s important to emphasize that despite the pause on the V3 project, the insights and research from the V3 work have not been in vain. They are now being redirected, where relevant, towards effective planning and ideation for V4. For example, our prior work on asset listing processes and portals is now informing our Permissionless Market Listings Framework, and our initial explorations into V3 Market Maker Incentive Distributions are significantly influencing our ideation for the proposed V4 Launch Incentives program.

Incentivizing Behavior

Continuing this discussion, it’s clear there are many thoughtful ideas and valid points raised. If our grant application is approved, we’ll delve deeply into this research and return to the community with a well-thought-out, transparent approach to launching incentive rewards. As seen on this thread, we value community feedback and are keen on working together with community members. Here are some additional thoughts regarding the points that have been brought up.

Balancing Deposits, Governance, and Staking and Trading Volumes

Undoubtedly, trading volumes take precedence as a crucial metric among the four. This priority is mirrored in the allocation of incentives for each behavior. Acknowledging the formidable task of transitioning to V4 is imperative, especially with the shift to a new blockchain. Consequently, we emphasize not only the volume generation but also streamlining the initial steps (deposits) and ensuring the enduring vitality of volume through fostering a more committed and engaged user base (staking and governance).

Deposits Incentivization:
Incentivizing deposits boosts liquidity, which is crucial for a thriving trading environment, especially from scratch. The overall incentives allocated to deposits will be smaller than the incentives for trading volumes, but at the same time, it’s an essential pillar as it is the first hurdle as users transition to or join dYdX V4. We believe early bridging and depositing of liquidity should be rewarded despite potential Sybil attacks, which we’re devising strategies to counter.

Incentivizing governance aims to foster community engagement. Early adopters and active members should be recognized, although the reward weightage here will be small due to the challenge of gauging quality participation.

Staking elevates network security and trims trading fees. The incentive weightage is minor, but it’s exciting to foresee the community leveraging staking to fortify the dYdX chain and discover new utilities for the dYdX token.

Encouraging Retail Participation vs. Market Maker Activity

Both streams of activity hold considerable importance for our project’s sustained success. It’s well-recognized that platforms attracting substantial retail activity tend to draw in market makers and sophisticated participants naturally. The enticing fee rebates for high-volume makers play a pivotal role in this dynamic. Notably, referencing Market Maker programs in CEXes highlights the significant retail volume they attract, against which the proposed maker rebates stand as highly competitive.

As a result, a substantial portion of our efforts is focused on identifying and rewarding retail flow, in contrast to the naturally occurring informed flow. While the UI vs. API approach is interesting, we acknowledge its susceptibility to manipulation. Therefore, we are implementing additional techniques to discern retail flow accurately. With the aid of a grant, we plan to leverage clustering and classification methods to categorize different flow types and fine-tune incentives accordingly.


There have been some questions about a previous grant with Chaos mentioned in this thread. We’ve shared color on this previously, but will share the full context below.

Last year, the DGP had conversations with numerous dYdX stakeholders to understand funding priorities to have the highest impact. From these conversations, we prioritized three key areas that would help dYdX the most. This included funding the development of a new asset listing platform, market maker risk dashboard, and ongoing LP reward optimization research.

  • The new asset listing platform would empower the community to list and manage new markets on dYdX, growing volume and activity on the exchange.
  • The maker risk portal would allow the community to vet new liquidity providers and measure their performance for reward distribution, improving exchange liquidity, in turn attracting more users.
  • The LP reward optimization research would periodically review distributions to recommend changes and assess the effectiveness of these rewards, making community expenditures more efficient and improving protocol economics.

Chaos had extensive experience building similar tools for protocols like Aave, Benqi, and Uniswap. Having previously worked on dYdX projects, the team were already familiar enough to hit the ground running – allowing us to ship faster.

In May 2022, we signed an agreement with Chaos Labs to work on these deliverables. This agreement was dependent on the renewal of the grants program passing, so we opted to delay the funding announcement until v1.5 started. In the meantime, Chaos started to work on all three initiatives in order to meet the timelines.

By the time the grants program was renewed in August 2022, dYdX v4 had been announced. V4’s architecture is completely different from v3.

The migration meant most of our previous infrastructure and tooling grants built for dYdX v3 would no longer apply in v4, or would have to be completely re-architected to have an impact, including this project. With the v4 launch originally planned for later that year, the Trading team paused new listings on v3 and it became clear that the engineering lift to implement a community-led new asset program for v3 would not be worth the effort. The initial research done on LP rewards was productively used through a proposal to improve the formula, but again further changes were paused in favor of migrating the formula to v4. Similarly, the onboarding process and assessment of market makers was expected to change significantly with the launch of v4, making any new work potentially irrelevant as soon as v4 was launched. Ultimately, it became a risk that we would continue working on deliverables no longer applicable to the future of dYdX.

After exploring methods of repurposing this agreement, we agreed with Chaos that a termination in favor of new projects (with a different structure) would be favorable. Chaos still put in several months of work in order to meet timelines, and they are entitled to payment for the time they put in. The $300,000 is representative of the time and effort they put into achieving the outcomes of the original agreement, prior to the announcement of a new v4 chain.

With that said, both the DGP and Chaos are committed to making productive use of the previous deliverables. More importantly, much of the work they’ve done is now being used or repurposed for v4 related projects. The New Asset listing platform and Market Maker portal can be repurposed for v4, and are now also being leveraged for future projects. The LP reward research was used for a successful governance proposal, and continues to shape initiatives like this one.

This was a tricky situation to navigate - v4’s announcement was a surprise to all of us and it meant our previously planned year-long engagements needed to be re-scoped. We’re very happy with the work done by Chaos till now, and are very excited to see these new initiatives kick off.


Guys, now we have a signed agreement from 5th May 2022.

Seriously, what were you finalizing 5 months? after the contract was signed a year before @larry message

It’s impossible to verify anything with pdf file you provided

Announcing dYdX Chain
dydx v4 on cosmos was announced on 22 June 2022 for public audience. I bet the internal information about it was at least one month before.

Start date Aug 2022

I am not even commenting that Reverie signed an agreement for $100K per epoch subscription.

In my understanding, this is simply theft from the Treasury.

P.S. Maybe someone can move everything about $300K to a separate topic so we keep here only the discussions about V4 initiatives

1 Like

Thank you for the detailed explanations provided by both @chaoslabs and @carlbergman from the dYdX Grants Program. It’s commendable to witness such transparency and clarity from all parties involved. We appreciate the commitment shown by Chaos Labs in delivering value even amid changing dynamics. We’re optimistic about the future and eagerly await the fruition of the discussed initiatives. Your dedication to openness instills confidence, and we look forward to the continued growth and success of dYdX.

1 Like

I would like to close the topic on this $300K grant with screenshots from Discord.

It is widely understood as either a completely thoughtless business decision by the grant manager (Reverie) or a conspiracy aimed at appropriating the funds for Treasury. Everyone chooses their own version.

Chronology of Events

In April of this year, I discovered two transactions: one for 200K and another for 100K, which nobody knew about prior to April. They happened in May 2022 and December 2022. Hereafter, the explanations from Discord participants.

@carlbergman first explanation:

The community was waiting for exciting things.

@larry confirmation:

Additional explanation from @carlbergman

So, members of the community understood that the resources of @chaoslabs had been reserved, but force majeure circumstances compelled the postponement of work until further details on V4 were available.

More details from @Derek

User Momo asked the guestion:

Carl confirmed that:

If we look at the dydx grants website, it’s not the case anymore, and it’s a separate 40K grant right now.

I asked again to reveal the contract in May:

@larry said they are finalizing the agreement (what can you finalize if the contract was signed):

Additional info from @larry in reply to @Immutablelawyer

After reading all of the above, I again have the impression that the Chaos resources have been paid for and are waiting for their moment to create impressive products for V4.


What conclusions can be drawn from this situation:

  1. ChaosLabs performed well by securing an excellent contract. Credit is due to their competent sales and legal team.
  2. Reverie wasted $300K of community money.

I would like to conclude this with a famous quote by Larry:

And ask a question: What downside has Reverie wasting $300K of community money?

P.S Thank you @peary for making Reverie accountable


I would like to clarify that I am very impressed with the work of @chaoslabs , for example, in AAVE, as well as their offer to work for free for six months, their level of transparency, and their interaction with the AAVE community. The scope of work in their first contract was much larger with linear payment and bonuses for deliverables. The offer to AAVE was made in September 2022. At that time, they did not have the reputation they have now and had something to prove as a company.

Therefore, in this case, my grievances are solely directed towards Reverie, who clearly mishandled the situation as a client, withheld information from the community, and made promises about exceptional projects. dYdX received much less from Chaos Labs’ work than they could have received.

This is a lesson for the future; the more demands the community makes of service providers, the better products we will obtain for our ecosystem

1 Like

Well curated @chaoslabs

1 Like

Great to see community-focused initiatives aimed at incentivizing early adopters, facilitating user migration, and boosting protocol growth. Incentivizing desired behaviors and activities, including deposits, trading, staking, and governance, should result in a more vibrant and engaged early user community, a seamless transition to a v4 model, and the stimulation of trading volume.

It’s also encouraging to witness active discussions bringing numerous valuable points and contributing to a more transparent solution. However, there is still room for discussion around the formula being confidential and how to enhance transparency while mitigating risks by implementing some of the proposed ideas, such as introducing unpredictability, randomness, a live leaderboard, and bi-weekly reports.

1 Like

Hi @chaoslabs,

Now that the dYdX Chain is past the genesis launch and officially in the Alpha stage, we should consider how the dYdX community can encourage users to bridge their tokens onto the new chain and secure the network even though they can’t use the trading application and no staking rewards at this stage.

Does your program include the deposit incentives to be applied retrospectively from the staking at the Alpha stage or the incentives would be applied only after the incentives program launches?


As dYdX prepares for the launch of V4, @chaoslabs presents a compelling proposal. At Govmos (the governance arm of the PRO Delegators’ Validator), we recognize the challenges in transitioning and expanding the existing V3 user base and emphasizes the importance of liquidity and migration for the success of V4. The Launch Incentives program, supplementing the native V4 Rewards program, is designed to leverage the proven efficacy of Liquidity Mining and token reward programs in stimulating protocol growth and trading volume.

While the proposed Launch Incentives Program carries the noble goal of amplifying incentives to encourage traders’ migration to V4, it is essential to address potential concerns related to transparency. The success of any incentive program relies heavily on clear communication and openness about its structure, execution, and outcomes.

  • Lack of Transparency Risks: One potential risk lies in the lack of detailed information regarding the specific criteria for reward distribution and the mechanism for evaluating the program’s success. To ensure the community’s trust and understanding, it is crucial to provide transparent insights into how Chaos Labs plans to assess adoption rates, modify distribution quantities, and adjust the program’s duration as necessary.

  • Importance of Rewarding Legitimate Action: Another critical aspect is the necessity of distinguishing and rewarding legitimate user actions from potential abuse, particularly by automated bots. To prevent malicious actors from capturing an undue share of the incentive pool, implementing measures to verify and reward actions executed through the frontend becomes paramount. This not only safeguards the integrity of the incentive program but also ensures that early users genuinely contributing to the platform’s growth are duly recognized and rewarded.

In conclusion, while supporting the proposal for a Launch Incentives Program, it is crucial to address transparency concerns to maintain the community’s confidence. Regular reporting, clear communication, and a commitment to adapt based on real-time data will contribute to the success of the program. Moreover, implementing safeguards against potential abuse, such as rewarding actions through the frontend, will help ensure that the financial support intended for early users genuinely contributes to the growth and success of dYdX V4. Balancing transparency with proactive measures against abuse is key to fostering a fair and thriving ecosystem.

Thanks for reading ! If you like our job, you can support by delegating to PRO Delegators Validator


It is a well proposal, and looking forward to it.

1 Like

@chaoslabs has published a blog about the comprehensive overview of the program :pray:


Thanks to ChaosLabs for introducing their formula for v4 Launch Incentives. We want to use the opportunity to provide some insights into what we see as valuable governance activity from our experience in 20+ DAOs.

We want to help incentivize productive and aligned activity in a way that ensures governance activity is not abused for exploits

Valuable governance activity:

  • Writing proposals, that get passed
  • Voting
  • Publishing rationales for voting
  • Staking

Governance activity not producing direct value:

  • Replies to forum posts
  • Forum posts
  • Likes
  • Discord activity

The ultimate goal of governance activity is to get the DAO to do something. Thanks to the token-weighted voting process, any proposal that gets passed is valid governance output, at least in the regard that it is an expression of what token holders want.

Writing proposals and convincing the community to vote is hard work and necessary for the DAO to evolve. It should be incentivized, especially since incentives would encourage actors who don’t stand to benefit from their passing to write proposals.

Malicious proposals or scam proposals shouldn’t be incentivized and shouldn’t pass. We think governance attacks are a real threat, but we want to keep it simple for this reply.

Voting should also be incentivized because the amount of the active token supply in governance increases the cost of a governance attack, making the protocol more secure. We would advise paying out further incentives to those token holders who publish the rationales for their decisions, educating others, and contributing to the dialogue.

Staking increases the protocol’s security outright and is already incentivized through staking rewards. Should not enough of the token supply be staked, we think it might be sensible to add further incentives, with a sharp eye on what can be maintained over a more extended period.

For the “not valuable” category:

While we think replies, Discord activity, and likes are essential for a thriving governance ecosystem and good decisions, these activities are too easy to farm. Furthermore, the farming activity would endanger the quality of the well-meant and meaningful authentic discourse, negatively impacting honest actors.

We don’t see a straightforward way to incentivize this behavior that doesn’t have adverse second-order effects. One prominent example is MakerDAO’s SourceCred system, discontinued because the community found the signal to noise ratio to leave a lot to desire.

We advise not to incentivize this category.


Be careful on this, since there is a nuance to it. It can be quite easy to make a lot of posts indeed, but making posts with content which add value to the discussion is another story. That is still a lot of work, which requires following the threads closely and staying in contact with the community and the direction the project is taking. A validator (or governator for that matter) does not always need to start a proposal itself to still be valuable for the governance process in general. I even think it is the other way around; if the person doing the governance is rewarded for starting a lot of small topics which could / should also be covered in a more comprehensive approach. It also might lead to heavy sub-optimalisation by trying to create a lot of small meaningful threads, where sometimes taking a step back and looking at the bigger picture is better.

So, to make a long story short, it will heavily depend on the way the role is executed whether you should incentive reactions of new threads imo.


In an ideal world, that’s very true, but at the same time, it would be hard to keep track of token holders who properly do the above or just vote for the sake of seeking incentives. Do you have any idea of tracking those behaviors for proper incentive distributions?