Vectorbt

Vectorbt (pip install vectorbt) is a Python toolkit for trading research. It runs on NumPy and Numba, with an optional Rust engine. Indicators (vbt.MA.run, vbt.RSI, vbt.BBANDS and more) and signals run as arrays, so one call can sweep thousands of parameter combinations. vbt.Portfolio.from_signals then simulates them with fees and reports returns, drawdowns, trades and interactive charts.

What it adds to Moby. Fast parameter sweeps. Backtesting.py checks one rule on one token, and Vectorbt checks a whole grid of rules at once, like every fast/slow moving-average pair on a token's hourly candles.

How to pair it. Pull get_token_ohlcv for a Solana token, load it into a pandas DataFrame indexed by time (Moby returns candles newest-first, so reverse them), and run a crossover or RSI sweep. Use the result to shape a Moby strategy, not to pick a trade.

What it does not do. It doesn't trade or read Moby directly. Short, thin memecoin histories make sweeps overfit, so keep it to tokens with weeks of candles. Limit orders, leverage and parallelization are in the paid VectorBT PRO, not this open-source edition. Its license is Apache 2.0 with the Commons Clause: you can use, change and run it freely, but you may not sell it, or a product or service whose value comes mostly from it.