Dolly is a 12 billion parameter causal language model released by Databricks. It is derived from EleutherAI’s Pythia-12b foundation model and fine-tuned on approximately 15,000 instruction-response records generated by Databricks employees. The model is trained on capability domains from the InstructGPT paper, including brainstorming, classification, closed QA, generation, information extraction, open QA, and summarization. It is licensed for commercial use and available on Hugging Face.