Technical profile and market coverage
Bender
uses Bybit
on 15m timeframe.
Sample capital: $1000.
Recommended minimum: $200.
Technical notes
Generation 3 of the Bender engine is an end-to-end reinforcement-learning system that learns its trading policy directly from market interaction. Inputs combine trend, momentum, volatility and volume-flow features across multiple timeframes (15-minute up to weekly) together with live position state. Execution happens inside a structural risk box: take-profit and stop-loss are defined in margin terms and applied identically in training and in live trading. Learning is continual but releases are gated — every candidate must pass a standardized out-of-sample evaluation with minimum activity and profitability requirements and outperform the incumbent, otherwise the previous policy is kept. This training-and-evaluation discipline is what keeps the bot's behavior reliable and consistent.
Covered pairs (4)
SOL/USDTAVAX/USDTBTC/USDTBNB/USDT
Full bot details
Bender is built to be predictable: typically 1-5 trades a day, each with a pre-set take-profit and stop-loss on margin, and leverage it self-adjusts between 2x and 8x. When a trade moves against it, it follows rules instead of emotions — either one limited, model-confirmed additional entry, or a controlled exit. The model keeps learning from fresh data, but a new version replaces the live one only after proving itself better on data it has never seen.
Reinforcement-learning futures bot for a basket of major coins (SOL, AVAX, BTC, BNB) with a steady cadence of a few trades per day. Every position carries a predefined take-profit and stop-loss, and a new model version replaces the current one only after proving it performs better on unseen data.
Supported leverage range: 2x to 8x.
The model warm-up uses 1000 initial candles.
Features
Generation-3 reinforcement-learning engine with continual training on fresh market data.Release gate: a new version goes live only by outperforming the current one on unseen data.Structural risk box: margin-based take-profit and stop-loss, identical in training and live.Steady cadence of 1-5 trades per day with two-sided (long/short) coverage.Adaptive 2-8x leverage matched to market conditions.Limited scale-in (DCA) entries, only with renewed model confirmation.