Hyperliquid vs Lighter for Trading Competitions

Hyperliquid and Lighter look like the same kind of venue from the outside: perpetuals-only DEXs where your wallet is your account and nobody has to link anything. Both are listed in the venues directory as perpetual markets, both are the only two venues on which Voltrade currently runs PnL challenges, and on both a competitor registers with a wallet and starts trading.

Underneath, they are the two most different integrations the platform operates. The difference is not the venue's UX — it is what each one is willing to tell an outside observer about a trader, and that determines what kind of competition each can host well.

The one-line difference

Hyperliquid publishes fills. Lighter does not.

Hyperliquid is an L1 with a fully on-chain order book, so a trader's fill history is public state. Voltrade reads it directly from the public /info endpoint — no key, no OAuth, no read-only credential anywhere in the path — because the Hyperliquid address is the EVM wallet you already connected.

Lighter's per-account trade history is auth-gated: /api/v1/trades requires a token signed by that account's own API key, and the public recentTrades endpoint returns only the last 100 trades per market with no pagination. Historical fills for an arbitrary account are simply not retrievable that way. So the Lighter integration had to be built out of what is public, which turned out to be two things: a wallet-to-account-index lookup, and a cumulative volume counter.

Everything else downstream follows from that single fact.

How each is scored

Hyperliquid's job runs at 5-minute cadence rather than the 20 minutes used for volume-only venues, because a PnL leaderboard moves with the mark price and not only with new fills. Unrealized PnL on an open position changes every block whether or not the trader does anything.

Lighter's volume path derives per-day volume from differences between two snapshots of a monotonic lifetime counter. That has a property worth appreciating: a missed sample costs nothing, because the counter catches up. It also has a real limitation — the counter is an all-markets aggregate, so it cannot answer "how much did this wallet trade in one specific market". For market-scoped competitions the sync switches to walking the block explorer's history API instead, which serves complete history for any account with no authentication at all.

Two implementation details on the Lighter side that affect fairness and are worth knowing as a competitor:

• Sub-accounts are summed, and deltas are computed per index first. Your indices are re-resolved every tick, so a sub-account created mid-competition contributes only its own delta rather than injecting its lifetime total into that day.
• A pre-start baseline pass runs over scheduled competitions starting within 48 hours. Without it, volume traded between your registration and your first snapshot would be folded into your own baseline and lost. The baseline is reconstructed at max(competition start, your registration), and the job refuses to seed a baseline it cannot justify rather than guessing.

Hyperliquid's corresponding sharp edge is the fill cap: the fills endpoint returns at most 2,000 per call. The job carries an explicit truncated flag, but a truncated fetch is not rejected — it is counted, and the partial window is scored anyway. That matters because pages ascend from the window start, so a truncated fetch is missing its latest days and looks exactly like "this trader stopped trading", which under-scores the most active competitors. Where the flag does bite is settlement: a final pass carrying any truncated or errored participant is deferred and retried, and only once the grace window expires does it settle regardless, marked DEGRADED. One permanently broken wallet cannot hold everyone else's payout hostage.

Markets

This is where the two diverge most visibly for a sponsor choosing a venue.

Hyperliquid is perpetuals on its own L1. The directory lists its top pairs as BTC-PERP, ETH-PERP, SOL-PERP and HYPE-PERP, with 8-hour funding intervals. It is a crypto-perp venue and a very liquid one.

Lighter is perpetuals on a zk-rollup L2, and its market list has gone well past crypto. The live list Voltrade's campaign builder reads currently returns 214 markets, including a large block of equity and pre-IPO perpetuals — names like AAPL, NVDA, GOOGL, META, MSTR, COIN, HOOD and PLTR alongside international listings a crypto trader has no other obvious on-chain route to.

If your competition concept is "trade the market people already trade", Hyperliquid. If it is "trade something they cannot get anywhere else on-chain", Lighter. That is the sponsor-facing version of the difference.

Fees

Take these as the directory's own catalog values, not as a live quote. The venues directory lists Hyperliquid at a base tier of 0.02% maker / 0.05% taker, annotated in our own catalog as a base tier that is volume-scaled and subject to change. Lighter carries no fee figures in the directory at all, so we do not quote any — check the venue.

The reason we link rather than freeze a number: on a volume competition, a competitor with real size pays materially less than any headline tier, and a stale number in a blog post is worse than no number.

Which format each suits

Both venues support PnL challenges — they are the only two that do — but the fit is not identical.

Hyperliquid is the cleanest PnL venue on the platform. Challenge scoring needs realized PnL, unrealized PnL on open positions, and cumulative volume, all per participant, all reconcilable. Public fills plus public positions give all three from one unauthenticated endpoint. The scoring engine then applies a flat-at-start gate, cumulative volume brackets, max-cap pro-rata on realized PnL, an unrealized snapshot pro-rata, and a minimum-volume eligibility check, and ranks by absolute total PnL. Participants who were open at the start and never went flat, or who finished below the volume minimum, stay visible but are excluded from payout at settlement.

Lighter does PnL too, by a longer route. Its challenge scoring is driven by the block-explorer history replay: realized PnL is reconstructed by average-cost replay over complete history and cross-checked against the venue's own reported position and entry price on every sync. A WebSocket fill listener runs alongside it, but only as a latency optimisation — the replay is the source of truth. It works, and it is verified against the venue every pass, but it is meaningfully more machinery than reading a public fills endpoint.

For pure volume competitions, Lighter's counter model is the more robust of the two, precisely because a cumulative counter cannot lose data to a pagination cap.

Choosing

Ask three questions in this order.

1. What are you scoring — turnover or profit? For profit, Hyperliquid is the shorter path. For turnover, either works; Lighter's counter model degrades more gracefully.
2. What do you want people trading? Crypto perps: both. Equity or pre-IPO perps: Lighter only.
3. Do you need per-market scoping? Hyperliquid scores from fills, so scoping is natural. Lighter's default counter path is all-markets and switches to explorer replay when you scope — which works, but is the more complex of its two modes.

Which lane you land in depends on the format. Volume competitions on both venues are admin-launched: permissionless creators, agents included, are restricted to on-chain pool tracking, so a venue volume competition goes through Voltrade rather than the self-serve wizard. PnL challenges are the exception — creation routes to the challenge path before the restricted-creator check ever runs, and Hyperliquid and Lighter are the two venues in the supported challenge set, so anyone can launch one on either. The launch page is where the volume conversation starts.

The single-venue write-ups go deeper on each: Hyperliquid competitions and Lighter competitions. If you are still deciding between formats rather than venues, volume vs PnL competitions is the piece to read first, and how DEX trading competitions work covers the shared mechanics.