Marquitta Ullum lyUI

My dad is sixty-two. He has a little BTC in his hands—it's something I convinced him to buy a few years ago. Last month, I took him through the entire TBV process. It was more troublesome than I expected, but the gains were different too.
At first I thought the difficulty would be technical. Turned out, it wasn’t. He’d already gotten used to concepts like private keys and addresses. What really stumped him was the sentence, “My coins are still in my own possession, but I can’t withdraw them for now.” I had to explain it three times before he finally accepted it. What he meant, in essence, was: “Then is it really mine?” I finally gave him the analogy of a time deposit—your money is yours, the bank hasn’t taken it, but if you withdraw early you have to follow the rules. Once I said it that way, he understood immediately, and he accepted it faster than I did.
The real hurdles are in two places. One is the withdrawal waiting period—he needs to know in advance that during this time he won’t be able to move anything. Otherwise, when the moment comes, he’ll definitely panic and will start wondering if he’s been scammed. The other is choosing the service provider—he has absolutely no ability to judge. I had to pick for him, and I also had to explain clearly what could be affected if that step is chosen incorrectly. Neither of these things is very noticeable in the existing interface. For people like us who need to bring our family members along to operate it, the burden is real.
What reassures him is also very clear. I showed him the address—his coins were indeed still on the Bitcoin mainnet, not turned into something else. The withdrawal conditions are written into the script, so anyone can verify. After he finished reading, he said something pretty interesting: “So no one can make decisions for me.” I think that sentence captures the point more accurately than any explanation of returns. He latched onto the most core idea of this design.
My conclusion is: for someone with a certain level of understanding, this is a relief; for a complete beginner, it adds burden. It replaces the cost of trust with the cost of understanding—and the understanding cost can’t be outsourced.
Next, I plan to write the whole process on a single sheet of paper for him. I’ll also record his redemption time from this round, to see whether it matches my own time.
@BabylonLabs_io $BABY #baby

I didn’t add to this losing trade to average down and smooth out the cost. Adding to a losing trade is the fastest way to turn a small mistake into a big one. Averaging down dilutes the cost; what it magnifies is risk—this is never a worthwhile deal. If you’re wrong, admit it and start over with the next trade. #TradFi晒单

Even if TBV runs perfectly, most BTC will not come to be pledged
Technical discussions often assume a premise: once custody risk disappears, idle BTC will flow into the collateral market at scale. But judging from the actual behavior of holders, the main obstacle is often not trust—it’s lack of motivation.
Long-term holders’ core desire is to do nothing. Any form of collateralization introduces liquidation risk, tax events, and operational burdens. The yields gained from those costs are often lower than what they psychologically require given the certainty of their principal.
The actual funding profiles that would arrive are quite specific: they have stablecoin financing needs, can accept the liquidation rules, need to explain control over assets to auditors, and cannot tolerate single-issuer risk. This pool does exist, but its boundaries are clear.
This also means TBV’s reasonable competitive target is not all BTC, but the portion of funds that cannot use wrapped assets because of compliance or risk controls. Estimating the potential market by total market cap will systematically overstate demand.
There is also an implicit threshold on the demand side: institutions need custodians, auditing tools, and risk-control systems to all support this vault structure. These integration cycles are measured in quarters, not something that automatically completes just because a product goes live.
Therefore, the progress of @BabylonLabs_io should not be judged solely by technical milestones. We should also look at how many custodians, how many lending/borrowing markets, and how many risk-control service providers have truly completed integration and moved into production environments.
I will track the actual size of pledged BTC, the concentration of its sources, and the average holding duration. The value of $BABY depends on how much real capital gets put into this system—not on how many people agree with its design approach.
@BabylonLabs_io $BABY #baby

I haven’t written anything about macro for a long time, but TBV reminded me of an old question: what ultimately supports Bitcoin’s security budget over the long run.
Block rewards are going down. This is written into the protocol—there’s no room for negotiation, and it won’t change just because the price rises. Over the long term, the system’s security will increasingly rely on other sources. Transaction fees are one part. Another part very likely comes from whether BTC itself can generate real economic value without leaving the mainnet. This isn’t an emotional issue, and it isn’t a narrative issue—it’s a mathematical one, and when the time comes, you naturally have to face it.
I think TBV’s significance lies right here. It doesn’t wrap BTC and send it somewhere else. Instead, it lets holders deploy their capital while keeping self-custody, providing guarantees for networks that need security, and receiving a return in the process. This path is fundamentally different from the old approach of “hand over coins to earn interest.” The underlying assumptions are completely different: the former doesn’t add custodians, so risks don’t stack; the latter adds another layer—each additional layer is another potential point of failure, and in recent years, every failure has come from exactly those layers. None of them escaped.
My doubts are also very concrete. How big is the demand side, really? How many networks are willing to pay for security long-term? At what level can the unit price be supported? Can this market grow into scale? We still don’t have long enough data to answer those questions. Any claims that talk only about aspirations but not about demand, I’ll discount. This is also the most important point I care about when looking at assets like $BABY —its value ultimately has to be supported by real paid demand, not by storytelling, and stories never last long enough to carry them.
But at least the direction is right. Bitcoin’s biggest waste isn’t price volatility—it’s that trillions in capital have long been left completely idle, and at the same time you can’t just hand over custody rights simply to make it move. For the past decade, basically no one has solved this problem well. TBV tries to address both of these things at once, and this line of thinking deserves to be taken seriously for a while.
Next, I plan to record changes on the demand side each quarter—only looking at data, not listening to stories.
@BabylonLabs_io $BABY #baby

To understand the value of TBV, you first need to see what it replaces.
Traditional Bitcoin cross-chain custodial solutions are generally threshold multisig: out of n signers, t signatures are enough to move funds. Security is therefore expressed as the phrase “at least n−t+1 people do not conspire.” The problem with this assumption is that it fails as the size of the conspiracy grows. But conspiracy is an off-chain behavior: it is invisible on-chain and cannot be proactively prevented. Historically, most bridges that went wrong failed here.
What TBV aims to do is to change the direction of the assumption: from “most people are honest” to “at least one person is honest.” The spending path of funds is fixed in advance by pre-signing. Whether the funds can be successfully recovered depends on whether the operator’s assertion is correct, and any single honest participant has the ability to submit a fraud proof that overturns it. To cause harm, everyone must be silent at the same time; it’s no longer a matter of reaching a threshold.
The difference along this direction is substantive. The former’s security is diluted as the number of participants increases; the latter’s security strengthens as the number of participants increases, because with one more person comes one more chance of being reported.
The costs are also clear. The challenges window, the liquidity prepayment mentioned in the earlier sections, the key assumptions behind the pre-signing ceremony, and the fee liveness—all of these are the added complexity paid to swap that assumption. Whether this complexity is worth it depends on whether it can run stably in real environments. I understand the mechanism design; what remains is to look at the data from actual deployments.
@BabylonLabs_io $BABY #baby

When studying TBV @BabylonLabs_io , I paid special attention to its peg-out process. The design of the fraud challenge period in this part is the backbone of the entire TBV security model, but it is also the cost most likely to be overlooked by users.
First, the mechanism. When a user wants to redeem assets into BTC, the counterparty will initiate a peg-out transaction. This transaction does not take effect immediately; instead, it enters a challenge period. During this period, any watcher can submit a fraud proof. If the counterparty tries to steal BTC or submits an incorrect state, the proof will be verified and the transaction will be blocked. Only after the challenge period ends with no disputes will the peg-out be finally confirmed.
This design essentially ties “security” to “time.” The longer the challenge period, the more opportunity watchers have to detect issues, but the longer users’ funds are locked. The shorter the challenge period, the better the user experience, but the narrower the window for fraud to be discovered. This is a standard trade-off problem, and Optimistic Rollup follows the same logic.
I think there are two details worth looking into. First, is the watcher role permissionless? If anyone can watch, anyone can submit challenges, and there is economic incentive for submitting challenges (for example, confiscating the malicious party’s staked collateral as a reward), then the security model is more solid. If watchers are whitelist-based or require special qualifications, then the risk concentrates. Second, how long exactly is the challenge period—seven days, fourteen days? This directly affects user experience and capital efficiency, and it also determines whether TBV can support DeFi strategies with short cycles.
TBV brings the trust assumption of BTC’s security level to cross-chain bridges, but the price is that users must accept this time cost. When evaluating product-market fit for $BABY , whether the challenge period can be accepted by DeFi users is the first practical hurdle—not a purely technical one.
@BabylonLabs_io $BABY #baby

I’ve spent a few years working on BTCFi-related things, and I’ve gone through every “bring Bitcoin out” scheme on the market—wBTC, tBTC, renBTC, and all kinds of LP bridges. After reading the TBV whitepaper seriously yesterday, I got pretty upset.
First, the wBTC setup: BitGo custody. Users send BTC to the custodian, and ERC-20 tokens are minted 1:1 on-chain. The risk model is simple and brutal—fully trust BitGo not to run away, not to misuse funds, and not to get frozen by regulators. This is pure centralized custody.
tBTC v2 is better: it uses a set of signers to manage BTC via tECDSA threshold signatures. The signers need to stake T tokens as economic collateral, so in theory, wrongdoing can be punished. But the BTC pool is centralized—once the threshold is broken, the entire batch of BTC is at risk. Also, users rely on the signers’ collective availability to process withdrawals.
TBV is completely different. It doesn’t “bring BTC out” at all. The BTC stays on the mainnet in its own Taproot UTXOs the whole time. The user is always one of the co-signers of the UTXO. The Covenant Committee only has approval rights over the pre-signed spending path; it can’t move the coins by itself. Even if the entire Babylon ecosystem disappears tomorrow, users can retrieve their funds independently after the unbonding timelock.
One sentence to summarize the difference: wBTC is “the custodian has BTC; you have an IOU,” tBTC is “the threshold bridge has BTC; you have wrapped tokens,” and TBV is “you always have BTC—you’ve just made a usage-right commitment.”
This difference is crucial for institutional entry. For compliance-focused funds buying BTC, the hardest hurdle is auditing—once funds leave a self-custody wallet, it triggers a whole bunch of processes. Under the TBV model, on-chain audit tools can directly scan the UTXOs to prove holdings, with no cross-chain or custody involved.
What do you all think about this “staking-in-place” vs. traditional cross-chain bridge battle—will it, in the long run, squeeze the share of wrapped BTC?
@BabylonLabs_io $BABY #baby

While researching Babylon’s token model, I noticed a two-layer structure: BABY staking and BTC staking coexist. Both participate in protocol security, but their roles are different.
BABY stakers mainly handle the protocol’s own PoS consensus, validating transactions and state transitions on the Babylon chain. BTC stakers, on the other hand, economically commit BTC to the protocol through time-locking, providing additional economic security for cross-chain verification and the finality of TBV. Both groups receive BABY incentives, but the boundaries of responsibility are clearly defined.
What’s interesting is the inflation allocation ratio. Of BABY token inflation, part goes to BABY stakers, part to BTC stakers, and part to protocol development and ecosystem building. This allocation model acknowledges a reality: BTC holders are the largest user group in Babylon’s TBV system, and their participation directly determines the ceiling of the protocol’s TVL. Rewarding them with real economic incentives is far more effective than just promoting a narrative.
From a game-theory perspective, this dual-staking design effectively binds the interests of holders of two different assets. BTC stakers want TBV to be secure, usable, and to connect with more DeFi use cases, because only then can their returns remain sustainable. BABY stakers want protocol usage to grow, so that fee revenue and token demand will increase as well. The goals of both sides are highly aligned, creating a positive feedback loop.
Compared with other BTCFi projects, many either rely entirely on BTC’s brand endorsement without real economic binding, or they use an independent token logic that has no relation to BTC. Babylon, through TBV, achieves functional-level coupling between the asset properties of BTC and the governance properties of BABY. This architectural difference is likely to become evident over the long term.
I can’t predict BABY’s short-term price, but I can say that if the adoption curve for TBV plays out, BABY’s value-capture model is clear and sustainable—not driven by sentiment, but by protocol cash flows and real usage.
@BabylonLabs_io $BABY #baby

EOTS is the hardest-core piece of the Babylon tech stack—I've been chewing on it for three straight nights just to barely get it down. Full name: Extractable One-Time Signature, meaning an extractable one-time signature. TBV needs it to implement slashing on the BTC chain.
The rough idea: a signer can only sign a given message once. If they sign the same message twice with different contents, the mathematical relationship between the signatures automatically leaks the private key. The whole process happens purely at the cryptographic layer—no additional arbitrators needed. It sounds like a fairy tale, but academia has studied it for many years, and Babylon has engineered it into a form verifiable by BTC scripts.
What does it mean for TBV? If a finality provider signs on two conflicting blocks, their EOTS signatures for the BTC staking pool would reveal the private key at the same time. Anyone who finds that private key can execute a slashing transaction on their behalf, moving their BTC away. The punishment doesn’t require committee decisions or governance votes—it’s triggered purely by cryptography.
A quick analogy: ordinary default penalties are like "sign contract → default → court judgment → forced enforcement." EOTS is more like "while signing the contract, the key is already left on the doorstep—at the exact moment you default, the key automatically flies out. Whoever sees it can open the door." The immediacy and unavoidability of enforcement are guaranteed by mathematics.
But engineering has boundaries. Section 14 of the whitepaper admits: EOTS assumes the signer’s private key is protected by a single entity. If you split it using MPC, detecting double-signing requires additional mechanisms. And in production, finality providers commonly use MPC to improve availability—this opens up new attack surfaces.
One of the issues the governance layer—$BABY —will need to address in the future is how to reconcile the purely theoretical idea of EOTS with the real-world engineering constraints of MPC. This isn’t a cryptography problem; it’s a systems engineering problem.
My take: EOTS is the most beautiful cryptographic breakthrough for TBV, enabling BTC to support proactive punishment via slashing. But beautiful theory inevitably takes a hit when translated into engineering—don’t just transplant the security assumptions from academic papers directly into the operating environment.
As usual, DYOR. EOTS is Babylon’s moat—a glass bridge between academic theory and engineering practice? Tear it apart in the comments.
@BabylonLabs_io $BABY #baby

Any so-called "de-trust" system ultimately has to answer a question: who actually operates the infrastructure, and why they don't do evil. TBV is no exception.
Architecturally, TBV involves several key roles. The Vault Operator coordinates users' deposits, withdrawals, and status updates. Universal Challengers monitor fraudulent behavior and, when necessary, submit challenge proofs. The Proof Generator produces the zk proofs required for unlocking and settlement. If all three of these roles fail—or are controlled by the same party—then "de-trust" becomes a paper promise.
What I care most about is the incentive design for challengers. In optimistic Rollup models, challenger incentives have always been a longstanding problem: under normal conditions, most transactions are honest, so challengers rarely get a chance to act. But if fraud is truly caught, will the reward be enough to cover the long-term cost of waiting and watching? If not, rational challengers will exit, leaving only a few specialized institutions—centralization risk quietly returns.
The role of the Vault Operator also needs careful dissection. Can the operator refuse service? Can they disappear at a critical moment, forcing users onto a complicated and cumbersome emergency exit path? TBV's design should include a forced-exit mechanism so users can recover BTC even if the operator fails. But whether this process is user-friendly for ordinary users is another issue at a different level.
TBV's economic security is not a single cryptography problem—it is a system engineering challenge involving multiple parties in game theory. Protocol parameters, slashing/penalty rules, and admission thresholds: if any one of these is set incorrectly, it could tip the entire balance. I expect to see more detailed economic security models released, including cost-benefit calculations under different attack scenarios, and stress-test results under extreme market conditions.
Technical feasibility does not necessarily mean economic coherence. These two lines have to move together.
@BabylonLabs_io $BABY #baby

Stablecoins are now the most profitable track in the crypto world. Combined, the market caps of USDT and USDC exceed $200 billion. Circle and Tether both post annual profits in the tens of billions. In terms of collateral structure, the two major issuers rely mainly on U.S. Treasuries and cash. Decentralized alternatives like DAI/USDS are also shifting toward RWA. As the crypto asset with the largest total market cap, BTC has long failed to secure its rightful share in the stablecoin-collateral market; the root cause lies in custody risk. TBV @BabylonLabs_io has the potential to change this landscape.
TBV stands for Trustless Bitcoin Vault. BTC is locked into a vault contract on the Bitcoin mainnet that never migrates. Based on the BitVM3 design, complex computation is performed off-chain using garbled circuits, while only compact fraud proofs are kept on-chain. On the Ethereum side, cryptographically generated collateral-state credentials are produced for smart-contract calls. To unlock, you must submit the corresponding ZK proof of the contract state; liquidation also requires a ZK proof. There is no multisig, no custodians, and no oracles.
For stablecoin protocols, TBV provides a collateral option of “BTC-level liquidity + decentralized custody,” which has been almost nonexistent in the past. The whitepaper explicitly states that stablecoin minting is one of TBV’s core use cases. This implies that in the future there may emerge a BTC-backed decentralized stablecoin, where the collateral is self-custodied BTC, and minting and liquidation are driven entirely by ZK proofs. If this product can be made to work end-to-end, the market opportunity is not small.
The foundation is Babylon’s Bitcoin staking protocol, with TVL over $5 billion and more than 50,000 BTC locked. The first integration is Aave v4: lock BTC → mint a collateral credential → borrow stablecoins → repay to unlock. In May, the Gomining collaboration first introduced a real test with 1,000 BTC.
Let’s stay level-headed: BTC volatility is significantly higher than that of U.S. Treasuries. Using BTC as stablecoin collateral requires a more conservative collateralization-ratio design, which will reduce capital efficiency. The liquidation-delay risk under extreme market conditions also needs to be validated by a model. Still, the direction is sound—and worth watching.
@BabylonLabs_io $BABY #baby

There are mainly three paths for BTC into DeFi right now. The first is WBTC, which is custody by BitGo. It has the largest market cap and the best liquidity, but you have to trust BitGo as a company. After the change of management-control turmoil in 2024, trust has split badly. The second is tBTC, the multisig solution from Threshold Network. It’s more decentralized, but liquidity is thin. The third is TBV, given by @BabylonLabs_io —completely different in design: BTC never leaves the Bitcoin mainnet.
TBV’s mechanism is to lock BTC in a vault contract on the Bitcoin chain. Using the BitVM3方案, it generates verifiable collateral state proof credentials on the Ethereum side. Off-chain, it uses garbled-circuit circuits to handle complex computation; on-chain, only a minimal fraud proof is kept, keeping the costs under control. To unlock BTC, users must submit a ZK proof of the corresponding contract state. Any liquidator who wants to move the collateral must also submit a ZK proof. Throughout the process there’s no custodian, no multisig, and no oracle.
Compare them and it becomes very clear: WBTC trusts a custodian; tBTC trusts a set of signers; TBV trusts only cryptography and the challenge period. The first two are “human problems,” and the latter is a “math problem.” Math can have bugs fixed; human problems are hard to fix.
Babylon’s Bitcoin staking protocol is the foundation: it has TVL over $5 billion, with more than 50,000 BTC locked, providing TBV with a mature validator network and liquidity base. The first integrated use case is Aave v4: lock BTC → obtain credentials → borrow stablecoins → repay to unlock. The whitepaper also lays out extensions such as stablecoin minting, perpetual margin, and liquid staking. The May Gomining partnership was a real-gold test at a scale of 1,000 BTC.
No sugarcoating the downsides. TBV’s user experience is more complex than WBTC’s. The challenge period slows down the operation cadence, and the auditability of the off-chain circuits is also a new issue. In the short term, WBTC’s liquidity and convenience will still have the advantage. But if TBV can encapsulate the complexity to the point that users don’t perceive it, long-term replacement is not impossible.
@BabylonLabs_io $BABY #baby

After 2024, global regulation of crypto derivatives has clearly tightened. Against this backdrop, when re-examining the positioning of <@grvt_io >, I think the route it chose is actually forward-looking.
First, the background. The U.S. SEC and CFTC have increased the intensity of enforcement against contract-based products. The EU’s MiCA is now fully in effect. Compliance license frameworks in Hong Kong, Singapore, and Japan have been gradually improving. The traditional “offshore + anonymous” crypto trading model is being systematically squeezed. The choices users will face next are: either accept KYC on compliant platforms, or trade in gray areas that lack protection.
$BTC
GRVT has chosen a middle path that is friendly to compliance without sacrificing user sovereignty. It has a license (according to publicly available information, it is registered in Bermuda and holds the relevant approvals). It performs institutional-level KYC, while preserving a non-custodial nature—funds remain in the user’s smart account, so the exchange cannot freeze or misappropriate them. In an environment where regulation is getting stricter, this combination has more room to survive than a purely anonymous DEX, and it can respond better than a traditional CEX to the “custody risk” layer.
What are the practical impacts on ordinary users?
First, in the future, platforms that can smoothly handle deposits and withdrawals will almost certainly require KYC. Refusing KYC means fewer and fewer optional platforms. GRVT’s KYC process is relatively user-friendly—standard mainstream identity verification and address proof are usually enough, without overly collecting data.
Second, compliant platforms have a higher probability of survival. In recent years, many exchanges that were shut down by enforcement had long avoided regulation. Choosing compliance-friendly platforms generally offers better long-term protection for fund safety.
Third, compliance does not mean centralization. GRVT’s compliance mainly exists at the legal-entity level; technologically, it remains non-custodial on-chain. These two aspects can coexist—you don’t have to choose one over the other.
A reminder: no single platform should carry all the positions. This is something I emphasize repeatedly across many posts. Diversification means not only diversifying assets, but also diversifying platforms and regions. Non-custodial on-chain platforms, compliant CEXs, and cold wallets each have their own roles.
Tighter regulation isn’t a bad thing. It will weed out bad actors who have existed for a long time, leaving teams that take product seriously. This process is a net positive for users.
<@grvt_io > <#grvt >

I’ve been thinking about a question: how exactly does @NewtonProtocol ’s “intent-centric network” differ from the traditional transaction model?
I’ve gone through the documentation several times. In plain terms, intent is what the user expresses as “I want this result,” not “I want to execute this step.” Traditional transactions are like: I want to swap 100 USDC for ETH, using the 0.05% pool on Uniswap V3.
Intent is: I want to use 100 USDC to get as much ETH as possible, and the operator helps me compute the path.
That sounds like it saves users work, but I don’t think it’s that simple.
The intent model has a prerequisite: there must be a pool of operators willing to help users solve it, and the answers they produce must be better than what users could come up with on their own. Here there are two costs: one is the solving cost, and the other is the competition cost. Newton’s approach is to run operators inside a TEE, and simultaneously use policies to constrain the execution boundaries—so, in theory, it can handle both “help me compute quickly” and “don’t let you compute wildly.”
But I worked out the overhead at the experience layer. Before users sign an intent, they have to figure out what their policy is. If that policy is written too broadly, the operator might act in a gray area; if it’s too narrow, the intent can’t be executed. This trade-off is something ordinary users simply don’t know how to set.
My judgment is: the intent model is developer-friendly and curator-friendly, but not retail-user-friendly. If Newton wants retail users to use it directly, it needs an extra “foolproof” layer in between—some kind of template so users can click a few buttons and generate a reasonable policy. $BTC
Next, I’ll focus on when Newton can build that layer. No matter how elegant the technical architecture is, if users can’t understand it, it’s pointless. If that step can’t work, then an intent network will always be just a B2B toy. The real volume isn’t in the protocol itself—it’s in the packaging layer.
$NEWT @NewtonProtocol #Newt

#BinanceTurns9 1th anniversary, Binance has you—leave aside the tedious details and give yourself a little time to be alone. Curl up in your room and listen to music, tidy the cluttered desk, and let your mind go blank without overthinking. You don’t have to constantly meet everyone else’s expectations—your feelings matter most. Life holds countless gentlenesses; as long as you’re willing to look up, you’ll see starlight and an evening breeze. Live seriously, take care of yourself, and good luck will surely come running toward you.

As someone who occasionally runs a trading bot, when I look at a new trading platform, I usually start by reading its API documentation first, because many issues only become apparent once you actually plug it in.
I spent some time seriously studying the interface design of @grvt_io the other day, and I found that in some details it’s more thoughtful than I expected.
First, let’s talk about latency. Anyone who’s written a strategy knows that having unstable matching latency is even more troublesome than being overall slower, because backtesting and live trading won’t line up. GRVT handles matching in a high-performance off-chain environment, while still preserving deterministic settlement on-chain. This combination allows order response times to be close to those of traditional platforms, without having to endure the second-level confirmation typical of pure on-chain systems.
Next is the signing logic. It uses wallet signature authorization, so users don’t have to hand over API keys and the asset risk that comes with it, like they would on centralized platforms. After the strategy is running, even if a secret key is accidentally leaked, an attacker can’t directly transfer assets away because the withdrawal route is always locked to addresses controlled by the user themselves. This design is crucial for teams running multiple accounts or doing managed client investments.
It also didn’t cut corners on order types. Common types are all covered: limit, market, take-profit/stop-loss, Post Only, and Reduce Only. For grid trading or market-making strategies, the order amend/cancel frequency requirements are also well supported. What I care about most is cancel speed, and based on tests, it’s within an acceptable range.
Going a step deeper, it makes the clearing engine and risk-control logic relatively transparent. Margin calculation rules and the liquidation price algorithm come with clear documentation you can look up. This is very important for modeling during the backtesting phase. Many platforms hide parts of their liquidation mechanisms, which can cause strategies to suddenly fail under extreme market conditions. $BTC
Overall, GRVT leaves quantitative users more room than most on-chain derivatives projects. It doesn’t treat “on-chain” as a gimmick label—it puts the focus on the real goal: “actually being able to run strategies.” I think that pragmatic direction is quite rare.
@grvt_io #grvt

Last week I reviewed a batch of attestation records in the Explorer for @NewtonProtocol and calculated the average verification latency. The data isn’t bad, but it’s not good enough to support truly high-frequency scenarios.
Newton’s attestation process roughly goes like this: the agent submits a transaction request, the operator network verifies the policy, consensus is reached and the attestation is signed, and then the transaction is written on-chain. The main time cost of this flow is in the consensus portion of the operator network. Under the current mainnet beta, the average latency is in the range of a few seconds to a dozen-plus seconds, depending on the operator response speed and the consensus rules. $NEWT
For ordinary user scenarios, this latency is fine. But for agents doing arbitrage or MEV-related activities, it’s a fatal weakness. A delay of a few seconds means the arbitrage opportunity is already gone. Newton’s policy layer is suited for mid- to low-frequency automation, not high-frequency trading. This is determined by the architecture, not by parameters that can be tuned. #newt
More subtly, it’s how the latency changes as the operator network expands. In theory, the more operators there are, the stronger the security becomes, but the communication cost in the consensus stage also increases. If operators expand from the current dozens to several hundred, latency could rise to the order of dozens of seconds. That would significantly impact user experience. Newton needs to find a balance between operator scale and response speed. $SYN
My current assessment is this: Newton’s performance positioning is a trusted execution layer for mid- to low-frequency automation, not a platform for high-frequency trading. Once you understand this boundary, you can understand Newton’s application scenarios. Crossing that boundary with too high of expectations will lead to disappointment.
$NEWT @NewtonProtocol #Newt

An awkward reality for the on-chain perpetuals track is this: many platforms loudly claim to be “decentralized,” but there’s very little data that can actually be audited. @grvt_io makes me feel that this is worth writing about separately.
Let me first talk about a practical test I did. Last Wednesday, I opened a short ETH position on GRVT and got filled at a price of 3,428 with a size of 2.4 ETH. On the frontend, the trade showed as executed about 12 seconds after the fill. On the corresponding L2 block explorer, I found the settlement record: the matching result was packed into a batched transaction, which includes the order hash, execution price, size, and the taker/maker account addresses (anonymized). After that, the state root of this batch data is submitted back to Ethereum mainnet.
So while off-chain matching can’t be verified in real time, every executed trade can be fully audited after the fact. Anyone can just connect to the block explorer to verify whether their historical orders were truly filled and whether the execution price matches what the frontend shows. This capability is something most CEXs completely don’t have, and even many so-called “on-chain DEXs” only do it partially.
Things like the insurance fund, liquidation records, the ADL queue, and the flow of funds for Yield Layer are all on-chain and can be checked on GRVT. I pulled liquidation data from the past 30 days and did a quick summary: 1,247 liquidation triggers, 3 instances of positions going into liquidation (being undercollateralized), and all of them were covered by the insurance fund—there were no ADL triggers. For a platform that hasn’t been live for very long, this health metric is a plus.
It’s important to note, though, that transparency doesn’t mean zero risk. The fairness of off-chain matching still depends on the behavior of the matching engine. Even though the outcome can be verified afterward, if the engine maliciously injects an order or delays matching at a specific moment, users might not be able to detect it immediately when it happens. GRVT’s approach is to gradually transition toward a ZK proving architecture—so that the matching process itself can be cryptographically proven. Once that path is fully realized, the trust assumptions will further decrease.
In my view, a key measure of an on-chain project’s maturity is whether it can truly make “verifiability” concrete. GRVT’s current level of completion is in the first tier among on-chain Perps.
@grvt_io #grvt