VaultVision Research · Hyperliquid
Dashboard Blog @0xkayser

HLP vs LLP: Which Perp DEX Vault Wins Your Deposit?

Hyperliquid's $400M battle-tested HLP vs Lighter's $700M higher-yielding LLP. Honest side-by-side on yield, risk architecture, stress-test history, and where each one wins for real depositor capital in 2026.

April 20, 2026 9 min read by @0xkayser
TL;DR. LLP wins on paper — higher headline APY (~30% vs 10-25%), more TVL (~$700M vs ~$400M), newer segregated-strategy architecture covering crypto, FX, and RWAs. HLP wins on what matters when things go wrong — 2+ years live, survived JELLY and FARTCOIN without breaking, bounded drawdown behaviour documented on-chain. For most depositors: split 60-70% HLP / 30-40% LLP. Put 100% in LLP only if you've decided the higher yield justifies taking on a yet-untested tail-risk profile.

At a glance

HLP (Hyperliquid)LLP (Lighter)
TVL (Apr 2026)~$400M~$700-900M
Current APR / APY10-25% APR~30% APY (peaked 60%+ earlier)
Launched2023 (2+ years live)2024 (newer)
StructureMonolithic single-poolSegregated strategies (crypto / FX / RWA)
Tail-event recordJELLY (~$12M loss) + FARTCOIN survivedNo comparable documented tail event yet
Lockup4 days (fixed)Varies by strategy
RWA supportNoneYes (+ separate XLP pool)
Capital efficiency (as margin)No — locked for trading usePlanned — use LLP as margin (roadmap)
Host protocol market share~44% of decentralised perps~8-15%
TransparencyFull on-chain, real-time stateZK-proof verified execution

Big-picture read: LLP is a bigger, newer, higher-yielding vault with better architecture on paper. HLP is a smaller, older, lower-yielding vault with a documented ability to survive things breaking. Which of those descriptions you prioritise is the whole question.

What HLP and LLP actually do

Functionally, the two vaults run the same business model:

  1. Depositors pool capital into the vault.
  2. The vault uses that capital as market-making inventory against the protocol's orderbook — posting bids and asks, earning the spread on filled orders.
  3. The vault also serves as the liquidation backstop. When a trader gets liquidated at an unfavourable price, the vault absorbs the PnL (or gets paid, if the liquidation hits at a favourable price).
  4. Depositors earn a pro-rata share of the PnL after performance fees.

That's the common core. Every operational detail under it — how PnL is allocated, how inventory is split across markets, what happens in a tail event — is where HLP and LLP differ.

HLP specifics

HLP runs as a single pool across all Hyperliquid markets — perps, and increasingly HIP-3 builder-deployed perps. Deposits are monolithic: your capital is exposed to whatever market HLP's strategy is quoting that day. HLP deploys across three sub-strategies (A, B, C) internally but depositors get one blended yield across everything. A 4-day lockup on withdrawals is the main anti-panic mechanism. Full mechanics in our HLP deep-dive.

LLP specifics

LLP runs a segregated-strategies model since the February 17, 2026 upgrade. Deposits route into buckets by asset class:

Risk parameters, liquidation waterfall, and ADL (auto-deleveraging) treatment are tuned per strategy. For RWA markets specifically — which have fundamentally different volatility and liquidity characteristics from crypto — Lighter created a separate XLP pool so the core LLP doesn't take direct RWA inventory exposure. That's a meaningful architectural improvement over a single-pool model.

Architecturally, LLP's design is ahead of HLP's. Segregated strategies + a dedicated RWA side-pool is what HLP will eventually need to build once HL's RWA markets mature. That doesn't automatically mean LLP is safer — newer architecture + less stress testing has its own risks — but on paper Lighter has shipped the more thoughtful 2026 vault design.

Yield: where it comes from, where it's going

Where the yield comes from

Both vaults generate yield from the same three revenue streams:

  1. Market-making spread. The PnL from buying at the bid and selling at the ask.
  2. Liquidation PnL. When a trader gets liquidated, the vault either absorbs the loss or captures the profit depending on the price at which the liquidation executes.
  3. Funding rate harvesting. When the market is heavily directional, funding payments flow from the losing side. Vaults sized correctly can capture funding.

The main difference is that LLP also earns Circle USDC revenue-share indirectly — Lighter's large USDC reserves generate T-bill interest, some of which feeds back into the ecosystem. HLP doesn't have an equivalent external income stream at comparable scale.

Why LLP's yield is higher today

Three factors:

The yield convergence is coming. HLP went from 40-60% APR in 2023 down to 10-25% today as competition arrived and TVL grew. LLP will compress the same way as serious market-makers arrive, institutional flow deepens, and TVL grows past the current level. Sustainable long-run yield for both vaults is probably in the 8-15% range. LLP's current ~30% is the early-stage bonus, not a steady state.

Risk architecture: monolithic vs segregated

HLP: monolithic, conservative, stress-tested

HLP's single-pool design is structurally simpler. One pool, one blended P&L, one withdrawal queue. The advantage is transparency and ease of reasoning — everyone knows exactly what HLP is exposed to. The disadvantage is concentration: when JELLY happened in March 2025, HLP took the full ~$12M hit, depositors felt it, and recovery took weeks of subsequent profits to offset.

HLP's mitigations have been practical: tighter OI limits, better liquidation parameters, more conservative position sizing. Not architectural — operational. That matters: HLP has kept its design philosophy consistent and adjusted parameters as it learned.

LLP: segregated, modern, untested at extremes

LLP's February 2026 upgrade is the most architecturally sophisticated vault design in perp DEX land today. Separating crypto / FX / RWA inventory means a tail event in one strategy doesn't automatically cascade into the others. Putting RWAs in a dedicated XLP pool means LLP core depositors aren't directly exposed to tokenized-commodity market risk at all.

On paper, this architecture should reduce depositor drawdowns relative to HLP's monolithic design. The catch: it hasn't been tested through a real tail event yet. Complex systems have emergent failure modes that don't show up in calm markets. When LLP has its first serious stress event — and it will — we'll learn whether the segregated design actually behaves as intended.

The asymmetric truth: on paper LLP's risk architecture is better than HLP's. In practice, HLP has demonstrated actual recovery from multiple tail events. Paper beats real only until real happens. For capital you care about, demonstrated-recovery is more valuable than architectural-elegance.

Track record and stress tests

HLP's documented tail events

LLP's documented tail events

This is the single biggest asymmetry between the two vaults. Depositors picking LLP are betting the segregated architecture works under real stress and that no whale-driven concentrated attack emerges in a way the strategy isolation can't contain. Depositors picking HLP have more data to go on about what happens when things break.

What each vault does uniquely

HLP's unique advantages

LLP's unique advantages

Allocation framework

A defensible split for a depositor with $X to allocate across perp DEX vaults:

Conservative (capital preservation-first): 80% HLP / 20% LLP. You accept lower yield in exchange for maximum battle-tested track record. The 20% in LLP captures upside and venue diversification without betting the farm on untested architecture.
Balanced: 60% HLP / 40% LLP. Default for most depositors. Split between demonstrated-recovery and higher-yield options, gets meaningful exposure to LLP's architectural upside without concentration risk.
Yield-first: 40% HLP / 60% LLP. You're pricing LLP's higher APY above HLP's longer track record. Defensible only if you've consciously decided the yield gap is worth the thinner stress-test history.
Not recommended: 100% LLP. Putting everything into the younger, less-tested option is asymmetric risk — the upside is marginally higher current yield, the downside is maximum exposure when LLP's first serious stress event hits. Do not do this with capital you care about.

Two additional notes:

Track every Hyperliquid vault, including HLP, live on VaultVision

Risk score, TVL, drawdown, depositor flow, and APR for every active HL vault. Free, no signup, updated every few minutes.

Open VaultVision

FAQ

HLP or LLP — which should I deposit into?

For most depositors: both, weighted 60-70% HLP / 30-40% LLP. HLP has the track record, LLP has the yield. Diversifying across both captures yield upside without concentrating risk in an untested architecture.

Why is LLP's APY so much higher than HLP's?

Earlier adoption stage (less competition per dollar of TVL), a wider RWA market-making opportunity, and a high volume-to-TVL ratio. Expect it to compress toward HLP's range over 12-24 months as LLP matures.

Is LLP safer because of segregated strategies?

On paper, yes — segregated risk is structurally more conservative than monolithic pool risk. In practice, segregated architecture hasn't been stress-tested through a real tail event yet. "Better architecture" is real but doesn't automatically translate to "better outcomes under stress" until we see the stress.

What's the minimum deposit for HLP / LLP?

No fixed minimum on either. Effective minimum is driven by gas costs — small deposits (under $100) get economically eaten by transaction fees. Practical minimum: $500-1000 for either vault.

Can I withdraw from HLP or LLP at any time?

Functionally no. HLP has a fixed 4-day withdrawal lockup — you queue a withdrawal, capital unlocks 4 days later. LLP's lockups vary by strategy bucket. Both structures are deliberate anti-panic mechanisms.

Did HLP survive JELLY and FARTCOIN?

Yes to both. JELLY cost HLP ~$12M in March 2025, FARTCOIN was a smaller similar-pattern incident. In each case HLP absorbed the hit, the team tightened parameters, and recovery happened within weeks. Full breakdown in our safety review.

Will Lighter ship LLP-as-margin in 2026?

It's on the roadmap as of early 2026 but no confirmed ship date. If launched, it's a real capital-efficiency feature HL doesn't match. Don't deposit based on the feature's expected launch — wait for it to be live.

Are HLP and LLP custody-safe?

Both are non-custodial in the sense that you interact via an on-chain deposit. Neither protocol team has direct custody over deposited funds. Both are subject to smart-contract risk, sequencer-risk (LLP, via Lighter's rollup), and validator-set risk (HLP, via HL's L1). Neither has had a custody-loss incident as of April 2026.

Related reading