Virtual AMM (vAMM)

A Virtual Automated Market Maker (vAMM) is a pricing mechanism used primarily in perpetual futures protocols where:

There are no real assets in the pool, but prices behave as if there were.

⸻

Core idea

A vAMM simulates an AMM like Uniswap, but: • No actual liquidity providers depositing tokens • No real reserves backing trades • Only virtual reserves used for price discovery

⸻

Intuition

In a normal AMM: • You trade against a real pool of assets

In a vAMM: • You trade against a mathematical curve • Your PnL is settled elsewhere (margin system)

⸻

The pricing model

Most vAMMs use the classic constant product formula:

x \cdot y = k

Where:

• x, y = virtual reserves
• k = constant

⸻

Example

Virtual reserves:

x = 1,000 (ETH)
y = 2,000,000 (USDC)

Price = y / x = 2000 USDC per ETH

If someone goes long (buys ETH exposure):

• x decreases
• y increases
• price goes up

⸻

Critical distinction

These reserves are:

• not withdrawable
• not real liquidity
• purely for price impact simulation

⸻

Architecture

A vAMM system has two major components:

1. Pricing Engine (vAMM)

• Maintains virtual reserves
• Calculates price impact
• Moves price along curve

2. Margin Engine

• Tracks trader positions
• Handles:
  • collateral
  • leverage
  • liquidations

⸻

Visual model

           Trader
             |
      (opens position)
             |
     -------------------
     |                 |
  vAMM            Margin Engine
(price curve)     (PnL + risk)

⸻

Where it’s used

The canonical example is Perpetual Protocol.

Also seen in:

• synthetic derivatives platforms
• early DeFi perpetual DEX designs

⸻

How trading works

Long position

• You “buy” from the vAMM
• Price moves up
• You profit if:
  • others push price higher
  • or external oracle confirms move

⸻

Short position

• You “sell” into the vAMM
• Price moves down

⸻

PnL

PnL is:

• not coming from pool assets
• settled via:
  • margin system
  • counterparty losses

⸻

Comparison: AMM vs vAMM

Aspect	AMM (e.g. Uniswap)	vAMM
Liquidity	Real tokens	Virtual
LPs	Yes	No
Slippage	Real	Simulated
Use case	Spot trading	Derivatives
Risk	Impermanent loss	Funding / liquidation
Backing	Fully collateralized pool	Margin system

⸻

The key mechanism: Funding Rate

Because vAMM price can drift from reality, you need:

Funding payments to anchor price to the real market

⸻

Concept

• If vAMM price > real price → longs pay shorts
• If vAMM price < real price → shorts pay longs

⸻

Why this matters

Without funding:

• vAMM becomes detached
• traders can exploit pricing

⸻

Deep insight

A vAMM is:

A price discovery engine without liquidity

This is fundamentally different from:

• order books → match buyers/sellers
• AMMs → use real reserves

Instead:

• vAMM → creates synthetic liquidity via math

Limitations of vAMMs

Be very clear here (most people gloss over this):

1. No real liquidity

   • Large trades can distort price easily

2. Requires strong oracle

   • Otherwise manipulable

3. Funding dependency

   • Core stability mechanism

4. Not capital efficient alone

   • Needs margin + insurance fund

⸻

Final mental model

• Segment Tree → efficient computation
• Liquidity Tree → execution decision space
• vAMM → synthetic price engine

⸻

Bottom line

A vAMM is a mathematical market maker that simulates liquidity instead of holding it, enabling leveraged trading without real pooled assets.