What is Inflexion

The first market to price Uniswap v3 in-range impermanent-loss risk on-chain — a collateralized, capped, bilateral convexity hedge with a pooled cvAMM floor and competing-MM ceiling.

Inflexion is the first market for Uniswap v3 impermanent-loss risk — and the first to price that risk on-chain.

A Uniswap v3 LP brings a specific position (its NFT), picks a duration, and pays a fixed upfront premium. At expiry the protocol pays the LP their realized impermanent loss — capped at MaxIL — trustlessly, from collateral that an underwriter locked up front. The product is precisely an in-range convexity hedge, not “IL insurance”: the cap is load-bearing, not cosmetic.

The payoff is $\min(\text{realized\_IL},\ \text{MaxIL})$. Entry requires the position to be in range — $P_a \le P_0 \le P_b$ is enforced at creation — and the cap is what makes full collateralization (and therefore the no-bad-debt guarantee) possible.

In one paragraph

If you’ve ever provided liquidity on Uniswap and ended up with less than if you’d just held — that’s impermanent loss. It costs LPs over a billion dollars a year and there’s no trustless way to hedge it. Inflexion is the first market that prices that risk on-chain: a pooled underwriter always quotes you a fair, code-capped price, market makers compete to beat it, and if your position suffers impermanent loss within its range you’re paid back — from pre-locked collateral, with no middleman who can run off with the money.

What it is, in one line

Inflexion is the first market for Uniswap LP convexity priced on-chain — MaxIL is the capital unit. A pooled cvAMM always quotes a code-capped fair price with dual-tranche structure (junior vol-seller, senior savings account), and competing market makers undercut it. The resulting flow is a structural DeFi volatility-risk-premium surface for Uniswap LPs — the first on-chain dataset of this market’s microstructure.

Why this market didn’t exist before

Uniswap v3 LPs carry a structural short-gamma exposure — their position value is a concave function of price, so they lose on both up and down moves relative to holding. Impermanent loss is the realization of that exposure, estimated at >$1B/year of realized losses across DeFi, with no trustless, non-inflationary way to hedge it.

Three things blocked a clean on-chain hedge until now:

No trustless instrument

Prior “IL protection” (Bancor v2.1) relied on token inflation to cover losses — a death-spiral mechanism, not a collateralized hedge. Inflexion mints nothing and pays from pre-locked USDC.

Beyond-range IL is unbounded

Past the range edge, IL grows without bound, so a product covering it could never be fully collateralized — and therefore never trustless. In-range IL is convex and bounded.

Two-sided cold start

A purely LP-vs-MM market has no liquidity until market makers arrive — a classic chicken-and-egg. The pooled cvAMM solves it by always quoting.

How it works, in five parts

1. Capped in-range IL transfer

   Coverage pays $\min(\text{realized\_IL},\ \text{MaxIL})$. MaxIL is the maximum in-range IL of the specific position — pure geometry computable at creation from $(P_a, P_b, L, P_0)$, frozen at creation, and identical across all three durations.

2. On-chain fair pricing

   The protocol computes and publishes $\text{FairPremium} = \text{fairRate} \cdot \text{MaxIL}$ on-chain via the FairValueOracle — an exact closed-form $\Phi$-sum, never a fitted surface.

3. Two rails, one settlement core

   Path A is the always-on pooled cvAMM (signature-free, on-chain). Path B is competing market makers (EIP-712 signed quotes). createSwapRouted gives the LP the cheaper of {pool, best MM quote}.

4. Full collateralization (FULL mode, the default)

   The underwriter locks exactly $\text{MaxIL}$. Since $\text{covered\_payoff} \le \text{MaxIL} = \text{collateral}$ by construction, FULL mode cannot produce bad debt under any price path — invariant I1, provable by Foundry invariant tests.

5. Non-custodial on-chain settlement

   At expiry, settle() reads the Chainlink round pinned at $T$, computes realized IL using the $L$ stored at creation, pays the LP $\min(\text{realized\_IL},\ \text{MaxIL})$ from the underwriter’s locked collateral, returns the residual, and returns the NFT.

The three pillars

The protocol stands on three pillars — read these before anything else.

On-chain published FairValue

An exact closed-form $\Phi$-sum, evaluated live per quote by the production Stylus FairValueOracle — never reimplemented off-chain. $\text{fairRate}$ carries all the vol/time dependence; MaxIL carries none.

The cvAMM — floor of liquidity

An always-on, signature-free pooled underwriter quoting off FairPremium, capped in code at $\text{FairPremium}\cdot(1+\text{maxLoad})$ by invariant I10. It removes cold-start: there is always a price.

MM competition — ceiling of price

Sophisticated MMs stream firm EIP-712 signed quotes (no last-look) and win only by beating the capped pool. Hedged MMs export short-gamma risk out of the system.

Floor + ceiling in one line: the cvAMM is the floor of liquidity (always quotes a code-capped price); the MMs are the ceiling of price (win only by beating it). createSwapRouted routes to the cheaper rail and never reverts on a bad MM quote — it falls back to the pool, and only the executed rail mutates nonce, capacity, and lock.

MaxIL — the cap and the unit of risk

MaxIL is two things at once, and both are load-bearing:

• The cap. $\text{IL}(P)$ is convex on $[P_a, P_b]$, so its in-range maximum is at a boundary: $\text{MaxIL} = \max\!\big(\text{IL}(P_a),\, \text{IL}(P_b)\big)$. Because $\text{collateral} = \text{MaxIL} \ge \text{payout}$, no bad debt in FULL is a construction, not a hope.
• The unit. MaxIL is pure geometry, frozen at creation, L-independent in the fair-rate sense — which makes positions fungible to an underwriter within a market. An MM quote is therefore per-market (a load + a MaxIL-ratio band + capacity), never per-NFT — the signed quote carries no tokenId.

See MaxIL — the cap and the unit of risk for the full doctrine, and the worked example for one position priced across three durations on a single, constant MaxIL.

Who it’s for — the four doors

Protect my LP position

Uniswap v3 LPs who want to hedge in-range IL — pay one number, get certainty. The prize customer is DAO / treasury protocol-owned liquidity that needs predictable, reportable P&L.

Earn in the cvAMM

Capital providers who want to earn the volatility-risk premium across two tranches: senior (a convexity savings account) or junior (a pure vol-selling, first-loss tranche).

Underwrite & compete

Sophisticated market makers who run their own vol models, stream firm signed quotes below the pool, and hedge short gamma off-system (Deribit / Panoptic / perps). Optional and additive.

Integrate & build on the data

Quants, vol-arb desks, and risk teams reading the data moat — the first on-chain view into the microstructure of the DeFi LP volatility-risk premium, exposed as free public SDK, subgraph, and REST API.

It’s live

Inflexion is deployed on Arbitrum Sepolia (chainId 421614), with a full create→settle lifecycle run on both rails on the 2026-06-05 stack — Path A paid an LP from the ConvexityVault, and Path B routed to a market maker whose own collateral was locked and paid out. Live addresses:

See the deployments reference for the full registry and the live status for the status of every component.

Structures ship from day one; the full DYNAMICS (pool-vs-MM spread as a forward-vol signal) mature with multiple competing MMs and volume. The dataset begins at the single redeploy; the off-chain telemetry halves begin at the first interaction logged.

What Inflexion offers today is the architecture and the first public view into the microstructure of the DeFi LP volatility-risk premium — not yet a mature dataset:

• Structures are present day-one. All five non-circular behavioral signals are defined and queryable from launch.
• Dynamics mature with volume. The richer signal — the pool-vs-MM spread read as a forward-vol indicator — only sharpens once multiple market makers compete and real flow accumulates.
• The dataset begins at the single redeploy, and the off-chain telemetry halves begin at the first interaction logged. History is served live from the subgraph via the API; if a backing (subgraph URL / RPC / telemetry) is absent, surfaces degrade to a typed pending state — never an error.

The honest line

Two claims, always separate, never merged:

1. LPs are always paid — no bad debt in FULL, code-enforced by invariant I1.
2. Depositors and MMs can lose principal — the pool is a volatility seller, and capital is not guaranteed for either tranche (junior is first-loss; senior is protected from underwriting loss only, never from the systemic tail).

LPs are always paid with no bad debt in FULL mode — qualified: capped payoff + solvent USDC + oracle/settlement liveness.

The no-bad-debt guarantee is exact only under the full clause:

• FULL collateralization — the underwriter has posted exactly MaxIL per swap, locked for the life of the position.
• Capped payoff — the payout is min(realized_IL, MaxIL). The cap is load-bearing: it is what makes payout ≤ collateral hold by construction.
• Solvent USDC — the collateral numeraire (dUSDC) holds its peg and remains redeemable.
• Oracle / settlement liveness — the price oracle stays live and settle() runs within the protocol’s settlement window.
• No rehypothecation breach — locked collateral stays instantly liquid; it is never routed to utilization-gated venues.

Stated without these qualifiers, the guarantee is not correct.

It is not Bancor (no token inflation), not GammaSwap (no active management), not Panoptic (the LP pays one number), and not “insurance” (it is a collateralized derivative). See what this is not for the full set of differentiators.

Where to go next

Quickstart

Cover a position, deposit, or pull data in a few lines.

Why Inflexion

The problem, the surplus, and the differentiators.

Why now

The 2025 theory (LVR, Lipton–Lucic–Sepp) and the Arbitrum + Stylus stack.

The math

The IL formula, the MaxIL derivation, and the $\Phi$-sum.

The protocol

Pricing stack, lifecycle, vaults, oracles, routing.

The data moat

Five behavioral signals into the DeFi LP vol-risk premium.