LLM playground to experiment with local models and build fine-tuning datasets and benchmarks - View it on GitHub