Models

← All models

Q-Learning (Baseline) (Lech Mazur baseline strategy)

Lech Mazur Benchmarks

Q-Learning (Baseline) is a non-LLM reference strategy in Lech Mazur's BAZAAR benchmark, not a separately released generative model or checkpoint. The official benchmark documentation describes a tabular Q-learning agent whose state is a discretized market summary and whose actions are different shading deltas; official logs identify it as qlearning.

Benchmark scores