The protocol utilizes the CALIBER system (Collateralized Asset Ledger: Insurance, Bailout, Evaluation, and Redemption) to tokenize physical hardware collateral into verifiable on-chain assets with enforceable property rights. It operates through bankruptcy-remote SPVs, equity down payments, and FiLo vaults that create multiple lines of defense for token holders, while managing illiquid collateral through market-driven mechanisms like QEV (Queue Extractable Value) rather than first-come-first-served redemptions.

The project operates an overcollateralized stablecoin (USDai) that generates yield by providing liquidity to emerging AI companies with real-world hardware collateral. Users can stake USDai to earn sUSDai, which provides real yield backed by AI infrastructure investments and can be used across DeFi protocols for additional returns.

USDai is a decentralized stablecoin protocol developed by Permian Labs that bridges traditional finance with AI infrastructure lending.

twitter

website

github

telegram

Real-world asset (RWA) issuers

USDai

# USDai Active loans Methodology

## 1. Overview

Active loans measures the total outstanding principal value of loans issued through USDai Stablecoin. It tracks the portion of the protocol's balance sheet that is currently lent out to borrowers, specifically representing interest-bearing debt. This metric is a key indicator of the protocol's utilization and its capacity to generate interest revenue from its lending operations.

## 2. Data sources

- **Blockchain events**: `LoanOriginated`, `LoanRepaid`, and `LoanLiquidated` events from `LoanRouter` contracts, covering Arbitrum
- **External data**: End-of-day price feeds and asset metadata

## 3. Step-by-step calculation

### Component A: Loan origination tracking

- **Purpose**: Captures the issuance of new debt and the initial principal amounts.
- **Logic**: `LoanOriginated` events are parsed from `LoanRouter` contracts to identify new loan positions. For each event, the principal amount, the currency asset address, and a unique loan identifier are extracted. These amounts represent the initial increase in the active loan book.
- **Key fields**: `principal` (loan amount), `currencyToken` (asset identification), and `loanTermsHash` (unique identifier).

### Component B: Principal reduction tracking

- **Purpose**: Captures repayments and liquidations that reduce the outstanding debt balance.
- **Logic**: `LoanRepaid` events are parsed to extract both scheduled principal payments and early prepayments, both of which are summed to determine the total reduction in debt for a specific loan. For `LoanLiquidated` events, the residual principal balance is calculated by subtracting the sum of all historical repayments from the original originated principal. This residual amount is then used to close out the remaining debt balance for the liquidated loan.
- **Key fields**: `principal`, `prepayment`, and `loanTermsHash`.

### Component C: Running balance calculation and USD conversion

- **Purpose**: Aggregates net changes into a daily outstanding value denominated in USD.
- **Logic**: Net daily changes are calculated for each unique loan identifier by subtracting repayments and liquidations from originations. A running cumulative sum is applied to these changes to determine the outstanding principal per loan. These balances are forward-filled across days to ensure the debt persists until a subsequent event occurs. The resulting daily balances are aggregated by asset, normalized by decimal precision, and multiplied by end-of-day price feeds to reach the final USD value.
- **Key fields**: `active_loans` (aggregated principal), `token_address` (asset identification), and `timestamp`.

*This methodology was automatically generated by Token Terminal's methodology analyzer.*

Value of loans currently outstanding to borrowers.

USDaiCHIP

Active loans

Active loans

Competitive landscape

Chain breakdown