EXTROPY-BASED
MARKET INTEL.
Markets move in probabilities, not predictions.
Hyperextropy delivers probabilistic market intel from thousands of AI-simulated futures.
Hyperextropy in Action
Market outlook from AI scenario distributions
FROM SIMULATION
TO LIVE DEPLOYMENT
The stack behind Hyperextropy and its deployment on Hyperliquid
THE STACK
Three layers working together
Frontier models simulating thousands of market scenarios per asset. No assumptions, pure probabilistic modeling from raw data.
Directional bias metric from scenario distributions. Computes a score (-100 to +100) by analyzing probability landscape. Positive = bullish skew, negative = bearish skew.
Experimental research deployment on Hyperliquid testing scenario-based allocation with transparent results. Educational purposes only.
Deployed on Hyperliquid
ABOUT HEP
Hyperextropy Provider: research in action
What is HEP?
HEP (Hyperextropy Provider) is an experimental research vault on Hyperliquid's permissionless infrastructure, forward-testing scenario-based allocation strategies using Hyperextropy scores to inform position sizing.
Applied research: testing scenario-driven allocation on-chain with transparent results, rather than just publishing theory.
KEY DETAILS
Educational Purpose: HEP is a research experiment, not an investment fund. All information for educational purposes only. See full disclaimer below.
LIVE VAULT PERFORMANCE
Real-time on-chain metrics. HEP research vault vs Hyperliquid's benchmark.
HEP VAULT
Hyperextropy Provider
APR
Loading...
TVL
Loading...
Max Drawdown
Loading...
Depositors
Loading...
Autonomous allocation based on scenario-driven forecasts. Voyons frontier models simulate thousands of scenarios. Rebalanced continuously. Strategy logic evolves as research progresses.
HLP VAULT
Hyperliquidity Provider
BENCHMARK
APR
Loading...
TVL
Loading...
Max Drawdown
Loading...
Depositors
Loading...
Hyperliquid's flagship liquidity provision vault. Baseline for passive index strategies on the platform.
RESEARCH VALIDATION
Before deploying on Hyperliquid's infrastructure, we backtested scenario-driven allocation strategies to validate the approach. Historical results below informed current research design. Past performance does not guarantee future results.
Feb 1 – Aug 1, 2025 (6 Months)
Out-of-sample backtest · 0.045% fees included
Backtest Methodology
Historical simulation using scenario-driven allocation. Position sizing proportional to scenario scores. TP/SL optimized via grid search. No traditional indicators used. Results informed research design (not predictive of live performance).
Test Conditions
6 months (Feb-Aug 2025) spanning bullish, bearish, and stationary regimes. Assets: BTC, ETH, BNB, DOGE, XLM. Fees: 0.045% per trade. Voyons Python SDK (voyons-tiny-25.9 backtest variant).
FOLLOW HEP
Ongoing probabilistic research deployed on-chain. Track live vault performance or build your own applications with Voyons.
Important Disclosure
Research & Educational Purpose: This page presents applied AI research in probabilistic forecasting. All content is provided for educational and research purposes only. HEP is a decentralized research vault, not an investment fund.
No Financial Advice: Nothing on this page constitutes financial, investment, trading, or other professional advice. Results shown are either hypothetical (backtests) or experimental (forward-testing). Past or simulated performance does not guarantee future results.
Third-Party Infrastructure: HEP operates independently on Hyperliquid's public decentralized infrastructure. Duon Labs has no formal partnership, affiliation, or endorsement from Hyperliquid. Duon Labs does not manage funds, hold custody, solicit participation, or control vault operations. All participation is voluntary and occurs entirely through Hyperliquid's platform.
Inherent Market Risks: Cryptocurrency markets are highly volatile and speculative. Any use of forecasting tools or market data carries inherent uncertainty. Users are solely responsible for their own research and decisions.
By accessing this page, you acknowledge that you understand these risks and limitations. For questions, contact contact@duonlabs.com.