TLP stands for Task Level Parallelism, that is the level of parallelism that BigML will use to perform a task that is decomposable into embarrassingly parallel tasks (e.g. building the models of a random decision forest). BigML offers five different levels. In the lowest, tasks are performed sequentially while in the highest level, up to sixteen tasks are performed in parallel. That is, the higher the level the faster your task will be finished. However, the more credits it will cost you.