{"repo":"Logarithm-Labs/fractal-defi","free":true,"listed":false,"github":"https://github.com/Logarithm-Labs/fractal-defi","clone":"git clone https://github.com/Logarithm-Labs/fractal-defi.git","description":"Open-source Python research library for DeFi strategies. Compose protocol-agnostic entities (lending, perps, DEX and LP) into typed strategies - backtest, simulate, track experiments.","language":"Python","stars":41,"topics":["backtesting","defi","python","uniswap","aave","ethereum","financial-engineering","hyperliquid"],"license":"BSD-3-Clause","category":"blockchain-web3","readme_excerpt":"Fractal Fractal — open-source Python research library for DeFi strategies. Compose protocol-agnostic entities (lending, perps, DEX and LP) into typed strategies; backtest, simulate, track experiments. Why Fractal Most DeFi backtesters are product-shaped: pick a protocol, run a strategy, get a P&L curve. Fractal is library-shaped — small primitives with a big composition surface. You write a strategy once against the generic BasePerpEntity / BaseLendingEntity / BasePoolEntity / BaseSpotEntity contracts and swap concrete implementations (Hyperliquid, Aave, Uniswap V3, GMX, your own) without touching the strategy code. Features - Protocol-agnostic entities. Concrete implementations for Aave V3 lending, Hyperliquid perps, Uniswap V2/V3 LP and spot, Lido stETH, plus generic Simple building blocks. Each entity is a typed state machine with update state for accruals and protocol-correct action methods. - Composable strategies. Register any number of entities under named slots, return ActionToTake from a single predict() hook, get a typed StrategyResult with metrics and a flat DataFrame. - Live + synthetic data. Loaders for Binance public REST, Hyperliquid info API, Aave V3 GraphQL, GMX, TheGraph (Uniswap V2/V3, Lido), plus log-normal GBM and bootstrap simulators for offline stress tests. - Experiment tracking. DefaultPipeline runs your strategy across a parameter grid, logs metrics + artifacts per run via MLflow, and supports sliding-window scenarios for stability analysis. - Type-s","default_branch":null,"files":null,"tree":[],"storefront":"/r/Logarithm-Labs","claimed":false,"request_supported":{"post":"https://gitbuyer.com/r/Logarithm-Labs/fractal-defi/request-supported","requests":0},"note":"indexed from public GitHub; nothing is for sale on this page. Clone it from GitHub. Paid listings live at /search."}