Tori：机构外汇市场15%收益率代币化，与加密货币不相关 | DeFi Frontier

At a very high level, Torus is a yield protocol that brings institutional-grade, delta-neutral strategies on-chain, starting with dollar-hedged carry trades in global markets. Mechanically, you swap USDC, USDT, or other accepted collateral for TRUSD on the protocol. TRUSD is backed one-to-one by USDC or USDT held in reserves, which the protocol then deploys across real-world fixed-income markets. The yield is generated from real economic activity, not from recycling capital into crypto.

Two things really set us apart from anything currently on-chain. First, the yields tend to be a lot higher than what you see in DeFi today, often in the high-single- to mid-teen range, and they’re not correlated with crypto cycles or crypto markets. Secondly, the whole thing is verifiable on-chain in real time through a third party called Accountable: reserves, liabilities, hedge ratios, everything. Even though the assets are held off-chain, there’s real-time visibility on-chain all the time.

Sam, thanks for joining us. How are you doing?

Hey, guys. Thank you for having me.

Part of the reason we met you in the first place is because you’ve been working on offering institutional yield to DeFi investors. We want to cover everything you’re building with Torus. We’re recording this in advance of Torus’ launch, so hopefully, for folks listening in, this is a great way to get caught up on what you’re building.

1. Sam’s background

That said, why don’t we talk a bit more about you as the founder, Sam, and tell us more about the team behind Torus?

Sure, happy to. I’m Sam, founder of Torus. Quick background about me: I got into crypto very early, actually, mining Bitcoin as a kid when it was still under $10. I wasn’t smart enough to hold any of it, so I’ll be making peace with that for the rest of my life.

Then, around 2017, I spent the summer in Zurich at Google, and my roommate was writing Solidity contracts for some companies, including insurance companies. That was my first real introduction to Ethereum and the whole DeFi space. The DeFi space was booming back then, but I wasn’t really into building in DeFi or crypto at that point.

For most of my career, my passion was software as a service. I worked at companies ranging from small startups to giants like Google, and then I built my own company starting in 2020. After 3 years, after COVID and everything, I sold it to the biggest exchange in Turkey, BtcTurk.

Afterward, even while I was building that company, after 2020 and 2021, I got serious about crypto. Since then, DeFi has been the center of my day-to-day—as an angel, as an early LP, and as a regular user at many protocols that you’ve probably seen in the space, from ones that fared miserably to giants we all know today, including many big yield products.

I’ve spent the better part of the last decade at the intersection of traditional finance and crypto. That’s what led to Torus.

Maybe it’s worth speaking about the team as well. The team behind Torus reflects the same kind of intersection. Most of us are either former founders from DeFi or TradFi—people who have actually shipped real products before—or come out of asset management and hedge funds. So, these are people who have run real institutional capital in real institutional markets, in global markets like Thailand, the Netherlands, Turkey, and elsewhere.

2. What is Tori? Institutional FX carry trades onchain

That’s honestly the mix you need to build something like Torus, because the product sits across 3 layers: product execution, real markets, and crypto-native composability. We need all the expertise we can get from those 3 intersections.

Sam, let’s shift into what Torus is. Let’s start really high-level: how do you like to explain this to people? What are you building at Torus?

At a very high level, Torus is a yield protocol that brings institutional-grade, delta-neutral strategies on-chain, starting with dollar-hedged carry trades in global markets. Mechanically, you swap USDC, USDT, or other accepted collateral for TRUSD on the protocol. TRUSD is backed one-to-one by USDC or USDT held in reserves, which the protocol then deploys across real-world fixed-income markets.

The yield is generated from real economic activity, not from recycling capital into crypto. Two things really set us apart from anything currently on-chain. First, the yields tend to be a lot higher than what you see in DeFi today, often in the high-single- to mid-teen range, and they’re not correlated with crypto cycles or crypto markets.

Secondly, the whole thing is verifiable on-chain in real time through a third party called Accountable: reserves, liabilities, hedge ratios, everything. Even though the assets are held off-chain, there’s real-time visibility on-chain all the time.

3. Why DeFi yields are falling short of investor expectations

Sam, I do want to get into the mechanics behind how your stablecoin product works, but let’s talk a bit more about the shortcomings of current stablecoin yields available to us as DeFi investors.

This is something I’ve also thought about a lot as a DeFi participant for the better part of the last decade. If you look at the current yield space, most of it falls into a few buckets, and each one has a ceiling or a shortcoming.

You have DeFi-native yield. You can lend on Aave, Morpho, and different markets. You can provide liquidity on Curve or Uniswap, or use other products like that. They have real products and real users, but the yield is mostly recycling crypto-native capital. There’s usually no net new capital coming in. When crypto is bullish, yields are great, obviously, but when it’s not, yields collapse. It’s basically a bull-market product.

Then you have T-bill-backed RWA stablecoins like Ondo. That’s real yield, and it’s audited, but it’s capped at the risk-free rate. You’re getting basically 4%, full stop, and you’re 100% dependent on U.S. monetary policy.

That may be interesting for treasuries or bigger players, but it’s not really interesting for everyday DeFi users like me who are looking for better yields than what DeFi has to offer, because that’s already accessible in TradFi. You can go to your bank account and already acquire those yields.

Then you have crypto-native delta-neutral products like Athena, Resolve, and many others. There’s also Structure, Sense, and Sensible[?]. The yield comes from perp funding rates, which are also highly cyclical. It’s the same issue as with crypto-native lending: in bull markets, it’s a great yield source. I’ve enjoyed that yield source for the better part of the last 3 or 4 years, but when the market is sideways or bearish, it’s not super great because you’re either at the risk-free rate or below it. You can’t really deploy into those perp funding rates.

Then there’s a newer bucket of yield products built around private credit or other illiquid RWA positions. These sometimes show attractive returns, but the trade-off is usually liquidity and single-strategy dependence. You may be locked in for months, quarters, or years, and your effective yield depends on execution, lockups, or secondary-market conditions.

Compared with that bucket, our goal at Torus is to offer liquid access to yield sources—something highly liquid, something that DeFi benefits very much from, with clear mechanics and a straightforward redemption and unwinding experience, which most of these products in this category cannot offer. I don't want to give any names, but obviously, you can see those in the market clearly.

What we think we can do better is actually pretty simple. One, give users access to a yield source that's structurally uncorrelated with crypto cycles. That's very important because it's driven by global money market monetary policies, not any kind of sentiment, crypto or otherwise. Two, make the entire backing verifiable on-chain in real time, so you don't have to take anyone's word for it, because most RWA products fall short on that.

Again, they have some reserves and put their name behind them, but it's not really visible in real time on-chain. And three, basically run it with institutional-grade execution, so you're not getting yield in exchange for taking some hidden tail risk. Again, we saw that fail in the past. I don't want to give any names, but last year, we saw a big one fail.

So obviously, we're not trying to compete with risk-free rates or stable rates, but we are trying to give people a different category: higher yield, high liquidity, real economic backing, and real transparency.

Yeah, I think that's a really good overview of what exists in the market today. I've been a pretty big proponent of bringing these exogenous yield sources on-chain because DeFi needs it, to be honest. I know that comes with risk, but there are ways to mitigate some of those risks, like you mentioned with Accountable.

4. How Tori strategies earn yield

Let's look a little bit deeper under the hood at how you're actually achieving some of these yields and what strategies you're employing. We've gone over the status quo of the industry, so what are you guys doing differently? Explain some of these strategies to us that you're going to be pulling off here with Torus.

The biggest basic strategy is carry trade in global money markets. Money markets, like federal funds markets and money markets in general, are probably something most people won't even know about. Money markets are the biggest piece of our portfolio.

The way to think about it is as a global carry trade. A carry trade, in its simplest form, is where you're borrowing a currency where rates are low, like the U.S. dollar, and lending or investing in a currency where rates are higher, usually some emerging-market currency like the Turkish lira, Egyptian pounds, or something else. Those are the main prominent ones these days.

The difference between those 2 rates is your yield. So the obvious follow-up is, okay, but what about FX risk? If you're running the trade unhedged, you're basically borrowing a currency where rates are low—the dollar—and lending in a currency where rates are high, like, say, Turkish lira or Egyptian pounds. You're exposed to the currency, right?

The currency exposure can get you down to 10%, 20%, or even 30% drawdowns in a single day. What we do differently at Torus is that we fully hedge the currency exposure back to USD. We have no open currency exposure. That's a hedged global money market carry trade.

The next obvious follow-up is, doesn't the cost of hedging eat the spread? In a textbook efficient market, like anywhere in Europe or the U.S., yes. That's covered interest rate parity, or CIP. That's very basic financial theory.

But in these emerging markets, the FX hedging market is structurally inefficient. That's because central banks actively manage the FX rate as a policy tool, and most local participants—institutions and retail—don't hedge. I've been a participant in these markets for the last 5 or 6 years, maybe more than that, and I never hedged my currency exposure because when you hedge your exposure, you're down to 8%, 9%, or 2%. Sometimes, even less.

But when you're running the trade unhedged, yes, you are taking more risk, which isn't really suitable for DeFi. As a normal trade for an investor, though, you're able to get 15%, 20%, or even 25% to 30% returns over a year. So the structural benefits of being the unhedged investor are much better.

People usually run the trade unhedged, so there is a shortage of buyers in the FX forwards market, the hedging instruments. The hedging market is uncrowded, and a meaningful spread still survives even after fully hedging back to USD.

In practice, we see anywhere from 3% to 10% or 15%, depending on the conditions and timing. That's by far the biggest sleeve in the portfolio. Those rates depend on different factors, like central bank policies and market conditions.

For example, when the Iran war started last month, rates went down from 8%, 9%, or 10% to 3% for a week or 2, and then they went up to 10% to 15% levels. It's a living, volatile market, but again, if the timing is right and everything is done well, you're able to get good returns for the risk you're taking, basically, in a highly liquid and highly scalable environment.

5. Capacity to scale the Tori trade + estimated yields

I remember when we first talked to you, Sam, and we were just learning a bit about Torus. One of the things that struck me—I don't remember the number—was just how much money could get involved in this trade.

Can you remind us what your estimates are for Torus in terms of how much money the protocol can take in to participate in this trade? And again, what is that base yield that we're expecting for folks earning yield through the trade through Torus?

That's a very nice question because that's one of the core reasons why we are going with a strategy like this. This strategy is highly liquid. It's been around for the better part of the last 50 years, maybe more than that, and it's being run by many of the world's biggest banks.

I don't want to give names, but there are certain banks running these trades in certain jurisdictions at levels of more than $30 billion, $40 billion, or $50 billion. So the capacity is very high, right?

The whole yield-bearing stablecoin space is what—maybe $25 billion these days, maybe even less than that because of the recent bank run, the one that was very horrific—but overall, this space is much bigger than that. The capacity it can take is immensely bigger than anything that you see on-chain.

The yield that you can get, again, comes from a volatile market. It can trade anywhere from 4% to 15%, but nowadays we see rates around 8%, 9%, or 10% at the collateral level. That's what the base collateral earns.

Because of the DeFi fractional-reserve mechanism, there's a 2-token model where there's a dollar product that earns more yield and is incentivized in other ways, and there's a staked product that gets all the yield. The base yield is then amplified by the staking ratio. If the underlying earns 8% and not every token is staked, that underlying can go up to 10%, 12%, or 15%.

That's how we're able to advertise numbers saying “up to X%” or “up to this percentage,” because that doesn't purely depend on the collateral yield. It also depends on DeFi market conditions, the staking ratio, and other factors.

6. Two-token model: trUSD vs strUSD

Okay, yeah. This model using 2 tokens is pretty familiar. For DeFi investors, we'll be accustomed to seeing TRUSD, which is a stablecoin, and then the staked version is where we actually capture the yield from the trade. Am I understanding all of that correctly?

All correct. Stablecoins are technically backed 1-to-1 by cash or cash-equivalent assets. Here, you're talking about a token backed by trading positions, so it's maybe better to call this a synthetic dollar—but apples and oranges, maybe. I'm not sure.

You have a dollar product that earns no yield, and you have a staked token that earns all the yield, correct?

7. Coming DeFi integrations for Tori

And then, Sam, what integrations might we anticipate for the 2 different tokens in terms of earning yield? Are you planning to create money markets or the ability to borrow against these with something like Morpho? Can we expect a market to launch on something like Pendle?

Obviously, yes. That's exactly why there's tokenization in the first place. You're bringing a carry trade from the global money market, earning X%, and putting it on-chain, where the borrow rates are actually less than that X%. So you're able to run a secondary carry trade in the money market on-chain.

Assume we're live on Morpho, Euler, or other lending markets, and assume the underlying earns 10%. If you're able to borrow at 6%, 5%, or whatever the current rates are, then you're able to loop that and get even better returns because of the DeFi dynamics.

That's something you're not able to easily do in traditional markets, by the way. That's one of the core value propositions that we offer even to traditional participants, because currently, with most money market instruments, you cannot go to a bank and get a loan against them, collateralize them, and basically leverage them.

The 8%, 9%, or 10% return you're getting in the traditional market is the final return you're getting. But here, in a crypto money market like Morpho, for example, you're able to collateralize the token. If you're working with the curator and the right partners, then you're able to collateralize it and use the funds for whatever operations you're running, or use the funds to loop the token again and basically amplify your returns.

So yeah, we are expecting a Morpho market to go live, a Pendle market to go live, and many other integrations, like a Curve pool and many others.

8. Risks to using Tori

Sam, I want to talk through some of the risks that might come with this kind of setup. I just pulled up your documentation while we're on here, and I like when I see people put the risks in their docs as opposed to glossing over them like they don't exist, because obviously there are going to be things that keep a founder up at night building in this industry. I think you're pretty deluded if you don't think that's the case.

Anyway, scrolling through this, you've got things like strategy risk, smart contract risk, counterparty risk, peg risk, liquidity risk, regulatory risk, and technology risk. I think you've really gone through the gamut of a lot of the surface area of worry, essentially. Maybe just walk us through how you guys are approaching risk in this space and managing other people's money, essentially, on-chain and off-chain.

Obviously, Torus is a hedge fund. It's not an asset manager. It's not managing other people's money when it comes to that, but overall, that's what many institutions do. Torus is not one of them.

When we go into risks, we have to be transparent with all of them. We're going to be able to mitigate some of them with certain things, and we're going to be able to offer some solutions to cover—or let people cover—some of the risks. For example, there are certain products on-chain that you can buy coverage or insurance against certain risks. We're already talking to a few of them, and we're going to be able to offer those to the participants and users. They will be able to buy them to mitigate some of the risks they see.

Strategy risk is obviously one of the biggest ones, but the big one in strategy risk is yield compression. Rates converging globally and the spread we're capturing shrinking—that's one of the main risks. There's a slow-burn risk, not a sudden one—not a collapse of the funds, not a drawdown scenario—but monetary policies change, and they don't change overnight. It moves over years, basically.

The other one in this bucket is currency dislocations, so think of a notable carry trade unwinding. For example, I guess, the 2018 Turkish lira event, or maybe 2015 in Swiss francs. These will happen, but in every one of those events, when you were running the trade hedged, you were fine. If you were running the trade unhedged, obviously, you took a drawdown.

That's the risk you're taking for the higher yield. When there's no risk, there's no yield, so you have to take some risks to get some yield. But your job as an investor, as a participant in any kind of protocol, is to assess the risks, assess the return, and make sure that the risk-reward ratio makes sense. What we're trying to do here is basically offer something with a very high risk-reward ratio, where the rewards are very high and the risk is minimal for the rewards you're taking.

If your principal is hedged, for example, in that carry-unwind scenario, the FX dislocation is a non-event for the principal because the principal is protected.

Another risk is counterparty and execution risk. This is a very real risk that is worth taking seriously. There's a clear distinction that we have to make. We are trusting execution desks, and in each market our execution runs through them. These desks have been around for the last 25, 26, 30 years, and they've already been managing tens of billions and hundreds of billions, depending on the desk. We are trusting them to execute our mandate properly, and we are trusting our FX-hedging counterparties. Those are the main 2 things that we trust.

As I've said, we work with regulated, in many cases state-backed, institutions with multidecade track records. First, one of them has 26 years of experience executing, basically, for pension funds. Second, our partners operate strictly within our mandate, so they cannot deviate from that. They execute what we say and nothing else. Third, we limit any single counterparty's exposure.

Fourth, on the FX-hedging side, the ultimate counterparty is the government plus major institutions in each local jurisdiction, like big banks that are already managing hundreds of billions of dollars, or global giants like JPMorgan and others that process trillions in flow every day. For the hedge to fail, every major institution in that FX market would need to blow up at the same time, because you can't take that instrument out elsewhere and then basically utilize it there. So the counterparty and execution risks can be mitigated in many ways.

There's also asset risk, I guess, because it's a synthetic asset. Obviously, you're dealing with the token, and there can be failures in execution, like an oracle failure, for example, which we've seen firsthand a few weeks ago. The lending-market integrations—everything you're touching in DeFi—there's a risk surface.

Our contracts are audited by the Sherlock and Nethermind teams. Both are public. We have a least-privilege role architecture, so every role can do only so much. All upgrades and every parameter change on the contracts are behind 24-hour timelocks.

No one, literally no one, even if someone gets all the private keys to all the multisigs, is able to mint unbacked tokens. That's a very big thing because that's again what we've seen in the past, very recently. If the team is able to do something, then that is usually exploited at some time. It's not a question of if; it's a question of when, right?

In every system, no matter how secure it is, it can get hacked. Your job as a founder, your job as a protocol builder, is to make sure that every risk surface is minimized. Even if something happens, the blast radius needs to be minimized as well. Overall, you're trying to secure everything, but if something fails, it's your job to make sure that the blast radius is as small as possible.

Overall, the crypto risks exist in every protocol. On the off-chain side, there are certain risks with execution that we also try to mitigate by, again, choosing our counterparties right and doing the execution right.

9. Expected launch time for Tori

Sam, before we wrap up, what can you tell us about the expected timeline to go live with Tori, and then any other milestones that we should be looking forward to in terms of getting access to Tori?

I guess this isn't public yet, but I can share that we're targeting going live within this month, within May. We're probably going to go live with a pre-launch vault, so we're hoping to share more details about that on our official channels very soon.

Beyond that, obviously, there are going to be integrations taking place both on-chain and off-chain. There are going to be partnerships that we announce. Overall, it's going to be an exciting few months ahead of us. We're going to share a lot in the upcoming weeks.

10. Closing

Very exciting. Again, we got to meet you just about a month ago, and we were really confident that this is the kind of product that the state of DeFi should be interested in. Unfortunately, there was this rsETH incident, and I feel like there have been a lot of calls for better, higher yields in order to risk having capital on-chain. I think if you can pull off this estimated yield through Tori, you're going to have a lot of demand for it.

Anyway, good luck with taking the protocol live. We'll be watching and excited to test it out. Sam, thank you so much for your time. Again, I really appreciate the product that you all are building, and we're excited to see it go live. I want to give you the final word here before you go.

Thank you so much for having me, man. It’s been a great pod. See you on the next one. Thanks, everyone, for tuning in.