I would like to start programming a multi task reinforcement learning model. For this, I need not just one maze or grid world (or just model-based), but many with different reward functions. So, I am wondering if exists a dataset or a generator for such thing, or do I need to code everything by my self?
Depending on your needs and the size of the project, you might be better off making a custom set of environments. If you'd rather not do that, though, you should take a look at OpenAI's CoinRun environment. A high-level description can be found in their blog post.
The "RandomMazes" version of this environment might be useful to you. And if you want to make the mazes smaller, you can redefine MAX_MAZE_DIFFICULTY in coinrun.cpp.
Note that, although the levels are procedurally generated, reward is only ever given when the agent picks up the single coin. So, this might not be as much variety in reward function as you wanted.