Summary
Kamino Lend is the largest DeFi protocol on Solana (excl. LSTs) with a market size around $2B. With its extreme focus on security and robustness, the protocol has earned the trust of Solana’s core DeFi audience. Now it’s time for the next evolution.
Today, we introduce our vision for Kamino Lend V2. Building on the battle-tested V1 codebase, Kamino Lend V2 enables the protocol to scale aggressively, while enhancing security, capital efficiency, and flexibility.
V2 will introduce a vastly improved borrowing and lending UX, alongside more sophisticated risk management infrastructure.
Kamino Lend V2 Overview:
- Market Layer
- Vault Layer
- Next-Gen Liquidation Engine
- Risk Management Architecture
- UX Automation
3. New Products
- Lending Orderbook
- Spot Leverage
Introduction
Why Build On V1
The V1 codebase has been battle-tested, undergone 10 external audits, and has been rigorously tested via internal smart contract testing. Our core users have come to trust the security and robustness of Kamino’s smart contract. Furthermore, Kamino Lend V1 has grown into a highly versatile and powerful lending platform—and we are confident that this infrastructure is mature enough to underpin the next evolution of lending on Solana. This codebase will remain the core primitive, providing a foundational layer on top of which v2 can be built.
Our Solana DeFi Vision
Solana rapidly expanded over the past year, with an explosion of new projects, assets, and users. As detailed in our Road to $10B proposal, however, we believe Solana’s breakout growth phase is far from over. We anticipate a continuous influx of institutional flows and retail users, as well as new waves of builders that take advantage of the scalability, and the growing liquidity and user base of Solana.
The vision for V2 is to establish Kamino as the financial layer that underpins this next phase of Solana DeFi’s growth. The foundation has been laid; the next wave of capital is coming, and Kamino will service it.
Kamino Lend V2
1. Core Primitives
1.1 Market layer
Kamino Lend V2 will embrace modularity, opening up the V1 infrastructure for permissionless market creation and enabling a diverse variety of markets to exist on the platform, serving different users, use cases, and risk profiles. Any market can be created with any combination of assets and unique risk configurations.
Creation and management of these markets will be available through the open-source, permissionless Kamino Manager SDK.
Features include:
- Unlimited asset combinations and vault configurations
- Risk oracles
- Collateral rehypothecation (yield on collateral)
- Customizable interest rate curve and fee parameters
- Automation features (looping, leverage, repay with collateral, collateral swaps, etc.)
- Margin calls, deleveraging
- Limit orders (target leverage, take profit, stop loss, deleverage, releverage)
1.2 Vault layer
With the growth of the Market Layer, yield opportunities will become increasingly diverse, making it much harder for users to manually optimize their yields. V2 introduces the Vault Layer: single-asset lending vaults that automate liquidity aggregation across various markets, providing optimized, risk-adjusted yields for every major token in Solana.
Vault Profiles
The Vault Layer will support virtually limitless customization. Each vault can deploy into a specified set of markets, with numerous triggers and risk parameters enabling autonomous or manual rebalancing.
A simple way to define these vaults are as Earn Accounts or DeFi Savings Accounts. However, instead of a one-size-fits-all approach, users can select a vault that optimizes for their yield expectation and risk profile, with the token of their choice.
Some examples of vault profiles can include:
- Risk-Adjusted: Maximize USDC yield, deploying only into markets with high-liquidity, low-volatility assets. Triggers: per-asset utilization caps, per-market allocation caps
- Yield Maximizer: Maximize USDC yield indiscriminately across all markets
- Thematic: Barbell portfolio, deploy 90% to low-risk and 10% to high-risk markets, or maximize SOL yield by deploying only into markets with LST-only collateral
Due to its flexible Vault infrastructure, V2 will also enable a brand-new product on Kamino: Orderbook Lending.
Partner Vaults
With the V2 vault infrastructure, Kamino is positioned to serve a previously untapped market in Solana DeFi: bespoke lending vaults tailored for projects, institutions, and fund managers.
These vaults can be seamlessly integrated into any external platform, providing users with direct access to Kamino’s yield opportunities while adhering to specific risk and liquidity criteria, and offering custom fee structures.
Several partner vaults are currently in the ideation phase, developed in collaboration with Kamino’s Risk Council. These vaults can be customized to meet virtually any requirements, and will have dedicated integration and product support. If you’re interested in creating and integrating a vault, reach out here.
Curator Vaults
With the introduction of the Vault Layer in V2, Kamino will also onboard external risk specialists, who will work in collaboration with Kamino’s Risk Council to curate these lending vaults, and ensure they are risk/reward optimized.
These curator vaults can significantly expand the vault layer, allowing for a far greater range of vault profiles to serve a greater variety of users and risk profiles. In turn, these specialists will share in the interest generated from their vaults. A variety of highly experienced risk specialists are already in the onboarding process, working closely with the Risk Council. Official introductions will follow in the coming weeks.
2. Infrastructure Upgrades
In Kamino’s product & development framework, nothing has more resources dedicated to it than protocol risk monitoring and smart contract risk assessment. We believe this is a primary reason why Kamino has earned so much trust in the ecosystem and become the lending protocol of choice for Solana users.
Building on the existing V1 infrastructure, Kamino Lend V2 will introduce a wide range of improvements:
2.1 Next-Gen Liquidation Engine
- Scam Wick Protection
- Limit Orders
- Liquidation Auctions
2.2 Risk Management Architecture
- Isolated Mode and Cross Mode
- Interest Rate Premiums
- Dynamic LTVs
- Auto-unstake
- Target Leverage
- Stop Loss/Take Profit
2.1 Next-Gen Liquidation Engine
Kamino’s approach to liquidation infrastructure is informed by two key principles:
- Lender solvency is paramount: Kamino’s liquidation engine should be extremely reliable, aiming to ensure lender (and protocol) solvency in the most adverse conditions
- Liquidations are bad UX: We aim to build products and mechanisms that protect users from liquidation and only perform liquidations if they threaten lender solvency
Both of these principles are reflected in Kamino Lend V1, with flagship products like Multiply and the JLP Market remaining liquidation-free since launch, and zero bad debt having accrued on the protocol since launch, largely due to its liquidation rules and risk parameters.
Kamino Lend V2 will add three key features:
Scam Wick Protection
Immense effort has gone into the development of robust oracle and pricing architecture on Kamino. However, sudden market movements can still cause user positions to be eligible for liquidation. Even if only for a fraction of a second, such a position could be liquidated. However, scam wicks can be identified and dismissed from liquidations, improving borrower UX significantly.
V2 will introduce scam wick protection, a mechanism designed to prevent liquidations in these scenarios while ensuring protocol safety.
Limit Orders
Limit orders are critical financial primitives that Solana’s low costs enable at scale. They are extremely versatile and can enable features like stop loss, take profit, target leverage, collateral swap, slippage-free position close, etc, allowing users to have more control over their positions, better capture the upside, limit downside, and avoid liquidations, etc.
Liquidation Auctions
Alongside the new limit order infrastructure, V2 also introduces a new, auction-based liquidation architecture to Kamino, creating a more competitive liquidation landscape that benefits both borrowers and the protocol.
In Kamino Lend V1, liquidation penalties are fixed between a minimum and a maximum value (eg. 5-10%). Liquidated users are then subject to a penalty of at least 5% and can go as high as 10% as position risk increases.
By contrast, V2 only specifies a starting penalty. This is the point at which the liquidation auction begins. Liquidators then bid to accept a reduced liquidation fee. With each improved bid, the liquidators agree to accept a smaller liquidation fee, in turn resulting in a smaller penalty for the user.
For Kamino, limit orders and liquidation auctions will enable a vastly improved liquidation experience, where the protocol and liquidators can lower the price impact losses incurred during liquidation while being less constrained by integration complexity. Ultimately, this allows more liquidators to be onboarded with better access to liquidity.
2.2 Risk Management Architecture
Due to the inherent volatility of crypto assets, as well as the liquidity constraints in Solana DeFi, risk parameters on Kamino Lend V1 have remained relatively conservative since launch. This has ensured the safety of the protocol and users but has also limited the growth of borrows in Kamino Lend V1.
Kamino Lend V2 will introduce three features that will further reduce solvency risk while empowering borrows to scale more aggressively than was possible in V1:
Isolated Mode and Cross Mode
Isolated mode will allow more fine-tuned intra-market risk configurations for specific collateral/debt combinations, allowing borrows to scale with significantly enhanced security and efficiency compared to V1.
Although V1 has a robust debt cap and LTV system, the surge in borrowing activity on Kamino has revealed the inefficiencies in the current system. In V1, debt caps are collateral-agnostic, meaning if an asset’s debt cap is raised, the additional capacity can be borrowed against any collateral in the market—operating on a first-come, first-served basis. In more purpose-built markets like the JLP Market, this is less cumbersome. However, in markets with numerous collateral and debt assets (eg. Main Market), borrows can only be scaled as far as the highest-risk collateral in the market allows.
By contrast, isolated mode allows for LTVs and borrows of a certain asset to be scaled aggressively against specific collateral types, whereas the global settings can remain static. Isolated mode can be as granular as needed—for example, a market could have specific caps and LTVs for every single collateral/debt combination.
Isolated debt caps are already audited and live in beta in the Main Market (labeled Isolated Caps), and have proven highly effective at scaling PYUSD borrows solely against SOL and wBTC collateral.
Interest Rate Premiums
Interest rate premiums are an elegant, dynamic interest rate system where borrow rates are not only determined by asset utilization, but are reactive to the risk of each borrow position based on its LTV and collateral composition. This is an additional feature of the Isolated mode where specific risk combinations can be set per collateral/debt pairing.
With the current utilization-based interest rate system, high-risk and low-risk debt positions are subject to the exact same borrow rates, whereas, in V2, position risk is priced into a user’s borrow rate via two mechanisms: Risk Premiums and Liquidity Premiums.
-
Risk Premium: Borrow rate premium based on position risk (LTV). As a position moves closer to liquidation, borrow rate increases, resulting in higher interest for lenders while encouraging the borrower to repay debt and reduce liquidation risk. Risk Premiums can be unique to certain markets or collateral/debt combinations.
-
Liquidity Premium: Borrow rate premium based on collateral composition. Every collateral asset on Kamino is assigned a risk tier according to Kamino’s Risk Assessment Framework. Collateralizing higher-risk assets will add a premium to the interest rate, thus pricing risk into positions’ borrow cost, and rewarding lenders for deploying liquidity against these assets.
Dynamic LTVs
In Kamino Lend V2, LTVs can be reactive to collateral price movements. Instead of a position’s LTV moving in tandem with collateral price movements (eg. 10% price increase leads to 10% LTV decrease), LTVs in V2 can be anchored to a certain collateral price point.
If a user is collateralizing SOL, for example, his Liquidation LTV could be 80% with SOL at $200. However, if SOL pumps 50% to $300, the user’s Liquidation LTV could reduce linearly by 50% as well, ending up at 30%. In effect, the user’s borrow power increases at a lower value than the SOL price increase, ensuring unnecessary risk does not enter the system. In contrast, the current LTV system simply increases borrow power in parallel with the collateral price increase, which can prove lethal during euphoric bull markets, where debt flows into the system at unsustainable levels.
Different collateral assets can also have different dilution factors, with some, for example not lifting borrow power at all with price increases. The parameters are extremely flexible, and can even vary between different collateral/debt combinations (eg. more USDC borrow capacity is safer than more WIF borrow capacity).
2.3 UX Automations
Kamino’s Multiply product has proven how valuable UX automation can be to users. With over $350m currently deployed into Multiply, the product’s automated yield loops, one-click leverage adjustment, and seamless collateral and debt management have provided a glimpse into what’s possible with the Kamino infrastructure.
Kamino Lend V2 will raise the bar, introducing significant improvements to Multiply, as well as introducing a brand new product suite for spot leverage.
Auto-Unstake
Since its launch, immense effort has gone into making Multiply as smooth and safe as possible, with features like stake-rate pricing completely redefining the risk landscape of leveraged staking on Kamino. V2 will introduce another step-change improvement to SOL Multiply: Auto-Unstaking.
One of the biggest UX limitations we’ve observed in Multiply is the price impact risk when closing a position. For large positions in particular, closing a position can incur substantial price impact, potentially eating into the yield earned via the product itself. Auto-unstaking solves this by allowing a user to close their Multiply position by directly unstaking the LST from the stake pool.
When closing a position, a user can simply opt to Unstake instead of Market Close. Instead of incurring slippage to instantly close their position on the market, the user simply waits until the end of the epoch for their LST to unstake. They receive their SOL, and the position is closed without any price impact.
Target Leverage
Some users may want to target always being 2x long on SOL or BTC. However, with the current tools, as price goes up, debt remains constant, and leverage decreases. Target Leverage can help grow exposure as the prices rise. Similarly, in the current lending model, as prices decrease, leverage will increase, pushing users closer to liquidation. Target Leverage will automatically deleverage a position, ensuring that it remains at the target LTV and does not reach liquidation levels.
Stop Loss/Take Profit
The new Spot Leverage product will also feature Stop Loss and Take Profit automation. These leverage trading tools will allow users to identify prices at which they wish their positions to be closed entirely—automating the workflow from start to finish.
3. New Products
Kamino Lend V2 will enable a huge range of new products on Solana, and to showcase what’s possible, Kamino itself will introduce two new products to the app:
3.1 Spot Leverage
DeFi Lending platforms play a significant part in enabling users to leverage their assets. Kamino Lend V2 will introduce a new looping product, Spot Leverage, which will feature:
- Optimized UI
- Vastly lower fees than perps
- Yield on collateral
- Soft liquidations
- Liquidation-protection features
Solana DeFi lacks a UX-optimized solution for longer-term leverage at more conservative leverage multiples (2-4x). This product is particularly necessary to large holders of SOL, BTC, or Memes, who simply want to increase their medium-to-long-term exposure. Perps platforms, for example, are often unaffordable for longer-term trades, and typically have extremely harsh liquidation conditions.
Kamino Lend has all the necessary smart contract infrastructure, risk management expertise, and deep liquidity reserves to bring a new, cost-efficient leverage product to Solana. Spot Leverage will be powered by Kamino flash loans, and will feature target leverage and stop loss/take profit mechanisms introduced above.
Significantly, the leverage product will tap into Kamino Lend’s pools, which means that, from day 1, users have access to a huge pool of existing liquidity. In addition, large users will not be subject to artificial position size limits. Whales can open positions are large as they wish, as long as there is enough liquidity available in the market.
We expect that Spot Leverage will service a core group of SOL, BTC, and Memecoin traders, and will grow into a widely used trading platform with a diverse range of trading pairs.
3.2 Lending Orderbook
Borrow and Lend orders are the final piece of the Kamino Lend V2 product suite, and will allow borrowers and lenders to express their intents fully instead of depending on a fixed pool formula. More implementation details will be revealed soon.
Conclusion
Ultimately, we envision a future where Kamino grows increasingly robust at its core, continually maturing its risk management, smart contract security, and liquidation engine. With an extremely secure foundation, the platform can then develop new infrastructure, products, and use cases.
Kamino Lend V2 aims to serve both of these needs. With new primitives to enable more products, improved security infrastructure, and novel UX improvements, Kamino Lend V2 is designed to service the next wave of on-chain activity on Solana.
The Kamino Lend V2 rollout is anticipated to start in Q4 2024, with a core group of vault managers already in the onboarding process, as well as a collection of flagship products currently in development alongside a variety of teams in the ecosystem.
We highly encourage you to leave your feedback on this proposal below.