Perp Gamma Exposure (GEX): Where Hyperliquid Market Makers Break

Price Tells You Where; Gamma Tells You What Happens Next

Two coins can sit at the same price with the same funding and behave completely differently on the next push. In one, a 2% move keeps going; in the other, it fades. The difference is who is on the other side of the book — and whether their hedging amplifies the move or absorbs it.

Options traders have a name for this: dealer gamma exposure (GEX). Short-gamma dealers chase price and accelerate moves; long-gamma dealers fade them. Perps have no options chain, so that lens has never existed for the perp market — until now.

The /api/v1/quant/gex endpoint builds a perp analog of GEX from Hyperliquid market-maker inventory and on-chain liquidation density, for every actively-traded coin.

What Is Perp Gamma Exposure?

Perps don't have literal options gamma, so we reconstruct the same behavior from where forced flow lives. We classify the full Hyperliquid account universe, isolate the market makers (the perp "dealers"), and map where their positions get liquidated. That yields three core fields per coin:

• mm_net_delta — dealer inventory: net long or short notional held by market makers.
• gamma_profile — the market-maker liquidation density by price: where dealer forced unwinds cluster. These are the short-gamma acceleration zones.
• gamma_flip — the price where dealer long-liquidations-below cross short-liquidations-above. The inflection between an amplifying and a dampening book.

Important: this is a behavioral analog, not an options dealer book. The market-maker set is a heuristic classification, so treat it as a strong positioning signal — not an exact gamma measurement.

The gamma_flip and dist_to_flip_pct

The single most actionable number is gamma_flip and your distance to it, dist_to_flip_pct:

Price below the flip with dense clusters just under mark means a push lower trips dealer stops that sell into the move — the classic short-gamma cascade. Above the flip, the book absorbs and mean-reverts.

Pulling the Gamma Profile for One Coin

It's a single Pro Plus call, filterable to one coin:

curl -H "X-API-Key: cdk_live_your_key" \
  "https://cryptodataapi.com/api/v1/quant/gex?symbol=BTC"

{
  "scope": "gamma_exposure",
  "coins": {
    "BTC": {
      "mm_net_delta": -48200000,
      "gamma_flip": 61850.0,
      "dist_to_flip_pct": -0.74,
      "regime": {
        "label": "amplify",
        "inputs": {
          "funding_skew": 0.013, "oi_change_pct": 4.1,
          "realized_liq_cascade": true, "dist_to_flip_pct": -0.74
        }
      }
    }
  }
}

Here BTC sits 0.74% below the flip with short-gamma dealers (mm_net_delta negative) and a live liquidation cascade — a book primed to follow through, not fade.

Amplify vs Dampen vs Transitional

The regime flag fuses the cluster profile with funding skew, OI rate-of-change, the realized-liquidation cascade signal and dist_to_flip_pct. Every input is returned in regime.inputs, so the call is reproducible.

• amplify — short-gamma analog: crowded book, dense clusters near mark. Forced flow follows through; momentum and breakout tactics fit, stops need room.
• dampen — balanced, mean-reverting book. Fades and range tactics fit; breakouts are suspect.
• transitional — sitting on the flip line. Highest uncertainty; size down until it resolves.

This is the same flag surfaced on the gamma exposure page and joined into the liquidation-map snapshot.

How AI Agents Use Perp GEX

Wire the flip distance and regime straight into tactic selection:

import httpx
g = httpx.get("https://cryptodataapi.com/api/v1/quant/gex?symbol=BTC",
              headers={"X-API-Key": "cdk_live_your_key"}).json()
btc = g["coins"]["BTC"]

if btc["regime"]["label"] == "amplify" and btc["dist_to_flip_pct"] < 0:
    tactic = "momentum_short"   # below flip, short-gamma -> follow-through
elif btc["regime"]["label"] == "dampen":
    tactic = "mean_revert"      # balanced book -> fade extremes
else:
    tactic = "stand_aside"      # transitional -> wait

• Tactic switch: momentum in amplify, fade in dampen.
• Stop placement: avoid parking stops inside dense gamma_profile clusters.
• Risk gate: cut size near the gamma_flip where direction is least certain.

Pair it with trader positioning for who is on each side and whale activity for conviction.