Community Support
Active community with over 500 members on Discord to help with fine-tuning questions.
Multi-GPU Support
Integrates with tools like xformer and Deepspeed for efficient multi-GPU usage.
Wide Model Support
Supports a variety of LLMs including HuggingFace, Cerebras, and more.
Fast Integration of Techniques
Quickly integrates new fine-tuning techniques for high performance.
Flexible Data Handling
Supports various dataset formats for fine-tuning.