Steer Protocol and Allora Labs partnered together to create a first of its kind and active, concentrated liquidity management strategy leveraging machine intelligence from the Allora Network.
The introduction of AI powered LP vaults heralds a new era in DeFi, where both institutional players and retail liquidity providers can benefit from:
- Enhanced capital efficiency: With more precise liquidity management, participants can expect higher returns on their investments.
- Minimized Risk: The AI-driven strategies significantly reduce the chance of incurring impermanent loss, allowing liquidity providers to engage more confidently in the markets.
The Limitations of Reactive Spot-Price-Based Strategies
Traditional LP strategies rely on spot-price triggers to rebalance positions. In a reactive model, the liquidity range is updated only after a significant price move has occurred. This lag creates several issues:
- Delayed Rebalancing: By the time the rebalancing action is taken, the market may have already moved further, exposing LPs to arbitrage and increasing LVR.
- Suboptimal Liquidity Deployment: Liquidity that is repositioned based solely on current prices may lag behind fast market moves, leaving capital stranded out-of-range.
- Heightened Impermanent Loss: When rebalancing occurs after significant price shifts, the asset ratios in the LP’s portfolio can deviate sharply from the optimal, deepening IL.
The Power Of Allora’s Predictive Price Feeds
The Allora Network provides AI-powered price feeds that forecast near-term movements for assets with exceptional accuracy. These predictive signals empower LP strategies to anticipate market moves rather than merely react to them providing benefits including:
- Preemptive Positioning: With foresight into where prices are heading, our strategies can adjust liquidity bands ahead of time. For example, if a price surge is predicted, the liquidity range can be shifted upward before the price actually moves, ensuring that capital remains in the fee-generating zone.
- Minimized Rebalancing Lag: By reducing the time between a market move and the liquidity adjustment, the integration effectively cuts down on LVR. Liquidity remains aligned with the active trading range, denying arbitrageurs the opportunity to exploit stale positions.
- Balanced Asset Ratios: Proactive rebalancing maintains a near-optimal portfolio mix, mitigating the risk of impermanent loss that typically arises from delayed reactions in volatile markets.
- Dynamic Liquidity Deployment: Predictive feeds allow for a nuanced adjustment of liquidity concentration. In stable conditions, a tighter range can be maintained for higher fee capture; during anticipated volatility, the range can be temporarily widened to reduce the risk of rapid out-of-range events.
In essence, Allora’s predictions equip liquidity management strategies with a “look-ahead” capability, ensuring capital is always deployed where market activity is about to occur.
The primary goal of the collaboration was to determine whether predictive insights truly lead to better outcomes for liquidity providers both in terms of performance and risk management.
The hypothesis of the collaboration was that Allora’s AI-powered price feeds would drastically reduce the time spent out of range by preempting significant price shifts ensuring LP tokens stay active in the fee-generating zone more often. The alpha derived from the price feeds could be used to mitigate Loss Versus Rebalancing (LVR) during volatile periods in the market.
Test Setup & Environment
Steer utilized a new data connector developed to fetch inferences from the Allora Network, specifically the 5 minute and 8 hour ETH price prediction topics. The testing period lasted from January 10 to January 28, comparing the traditional “Classic Rebalance” strategy with the new “Allora Strategy” (Allora AI Rebalance V1.2 Strategy, which is an enhanced version of the classic rebalance strategy). Both strategies were deployed under similar market conditions and tested on the deepest pools of ETH liquidity (USDC/WETH): Uniswap V3 on Arbitrum, DragonSwap on Sei, Camelot on Arbitrum, QuickSwap v3 on Polygon, SushiSwap on Base.
Test Results & Findings
Below, we share the findings from the raw performance numbers to the detailed breakdown of how each strategy behaved under different market conditions.
From January 10 to January 28, both the Allora-Integrated strategy (“Allora Strategy”) and Classic Rebalance strategy were monitored daily. As seen in the Performance Comparison chart:
- Allora Strategy (Purple): Averaged higher returns throughout most of the testing period, with occasional dips but consistently recovering more strongly.
- Classic Rebalance (Green): Performed well, but still lagged behind Allora’s performance line, with slightly more pronounced drawdowns during market volatility spikes.
LP Token Price Behavior
One of the most tangible indicators for liquidity providers is the value of their LP tokens over time.
In the LP Token Price chart above:
- Allora LP (Orange): Showed a steady upward trend from mid-January onward, reaching its apex around January 25 before a brief pullback. Even after the market dip, it remained at a higher price band compared to the Classic LP.
- Classic LP (Pink): Displayed similar movements but generally stayed at a lower price point, and its largest drop around January 27 was more pronounced, leaving it with a lower recovery level compared to Allora by the end of the period.
This confirms the hypothesis that predictive, proactive rebalancing helps LPs maintain in-range liquidity more effectively, thus preserving and growing the intrinsic value of their LP tokens.
The “Edge” Factor
The “Edge” (%) is a measure of the strategy’s advantage in terms of positioning and market alignment. The Edge Analysis chart often hovered between 0–5% for much of January, but we observed spikes to nearly 10% or more during times of rapid price movement. These surges underscore how preemptive positioning driven by Allora AI inferences can yield substantial, short-term advantages over reactive strategies.
Allora’s average performance ended in positive territory (+0.75%), while the Classic strategy’s average dipped just below zero (-0.08%). Moreover, the Allora approach achieved a significantly higher peak performance (4.77% vs. 1.86%) and also exhibited a slightly less severe drawdown on its worst day.
Conclusion: the Future of DeFAI Strategies with Allora
The partnership between Steer Protocol and Allora Labs has demonstrated the transformative potential of integrating predictive AI into DeFi liquidity management. The results speak for themselves—enhanced capital efficiency, minimized risk, and superior performance across volatile market conditions.
But this is just the beginning.
At Allora, we’re pioneering the future of DeFAI. By harnessing the power of predictive intelligence, we’re unlocking new dimensions of capital deployment, risk management, and yield optimization that were previously out of reach.
As the DeFi landscape continues to evolve, Allora’s mission is clear: to build the intelligence layer that empowers next-generation financial strategies, enabling both institutional and retail participants to thrive in an increasingly complex market.
The future of DeFAI is here, and Allora is leading the charge.
About Steer Protocol
Steer Protocol is a decentralized infrastructure platform that simplifies the creation, automation, and scaling of data-driven Web3 applications. By enabling seamless access to on-chain and off-chain data, Steer empowers developers to build sophisticated DeFi tools, automate smart contracts, and deploy omnichain applications without the need for centralized infrastructure.
Its high-performance network supports use cases like liquidity management, automated trading, and governance, offering unmatched efficiency and composability for Web3 innovations.
To learn more about Steer, visit the website, X, Discord and developer docs.
About the Allora Network
Allora is a self-improving decentralized AI network.
Allora enables applications to leverage smarter, more secure AI through a self-improving network of ML models. By combining innovations in crowdsourced intelligence, reinforcement learning, and regret minimization, Allora unlocks a vast new design space of applications at the intersection of crypto and AI.
To learn more about Allora Network, visit the
Allora website, X, Blog, Discord, and developer docs.
