Fragmented,
unpredictable
& constrained.
Fixed-rate borrowing doesn't need justification — it's the default in every mature credit market. The problem in DeFi is not why fixed rate is needed, but why it still doesn't exist properly at scale.
Fragmented Borrowing
Capital is scattered across multiple venues, forcing users to manually choose where to borrow.
- —Suboptimal pricing due to venue isolation
- —No unified view of available liquidity
- —Execution depends on user guesswork, not market competition
- VENUES4+
- PRICE SPREAD156bps
- MANUAL STEPS3–5
Unpredictable Cost
Floating rates fluctuate continuously, making it difficult to plan, manage risk, or run capital-efficient strategies.
- —No cost certainty for treasuries or strategies
- —Continuous monitoring required
- —Exposure to sudden rate spikes
- RATE RANGE2–18%
- 30D VOLHIGH
- PREDICTABILITYNONE
Capacity Constraints
Existing designs rely on isolated pools, AMMs, or constrained liquidity models, all of which limit capacity.
- —Limited borrowing size and inconsistent availability
- —Fragmented liquidity across markets and tenors
- —Pricing inefficiencies from pooled or standardized structures
- POOL DEPTHSHALLOW
- TENOR FLEXNONE
- SCALELIMITED
RFQ pricing,
aggregation
& management.
Built as an origination layer on top of existing onchain credit markets, IRIS is designed to enable competitive fixed-rate borrowing at scale. Among several key advancements, IRIS offers:
RFQ-based Pricing
Competitively priced fixed rate — tailored to your exact size, duration, and market conditions.
Multi-Venue Aggregation
Access deeper liquidity and more efficient pricing without relying on any single venue.
Active Liability Management
Optimized rates reflecting the global cost of capital without the constraints of a single static pool.
Intent to
settlement.
Assume your project name is 'Alpha' with aUSD (non-yield bearing) and saUSD (yield bearing).
Intent
The borrower specifies the exact loan parameters: collateral asset + amount, borrow asset + size, exact duration (flexible, not standardized), and maximum acceptable fixed rate.
RFQ
The request is broadcast to a network of solvers. Solvers evaluate the specific risk using proprietary models. Each submits a fixed-rate quote within a short time window.
Bidding
The best quote is selected. The winning solver commits bond capital to back the rate.
Settle
Borrower has a short window (~120s) to accept. If accepted: collateral is deposited, loan is executed on underlying venue, funds are delivered.
Manage
Add collateral anytime. Repay early (coupon preserved).
Safeguards
Re-auction (rare): only triggered under extreme solver insolvency. Bond-backed system: solvers are economically aligned to avoid failure. No forced liquidation at maturity: positions transition to penalty mode instead.
Research-backed
performance.
Multi-venue research across real market conditions designed to reflect how IRIS would behave in practice.
33.4%
In about one-third of recent planning windows, locking beat floating in hindsight.
2025–2026 planning benchmark; searched fixed quote vs realized floating over matched windows.
+125–200bps
125bps aggressive — 200bps conservative baseline.
At 125bps, bond-loss frequency is ~10.8%. At 200bps, it drops to ~4.7%, making it the safe positive baseline. Solver risk tolerance determines where within the range a quote lands.
156bps
The chosen quote beat the runner-up venue by about 156bps on average.
30-day searched borrower grid; this is the clearest 'search matters' number.
3.3%
0% in ordinary windows; 3.3% in late-2024 holdout.
Re-auction triggers at 95% of bond loss. 'Ordinary' refers to recent out-of-sample data; 'Late-2024 holdout' is a separate hostile-rate stress test period.