Zihao-Wu06/distillops
Black-box LLM distillation for any verifiable task, with text or image inputs: teacher API -> rejection sampling -> SFT -> eval -> GRPO
GitHub repository with 101 stars and 2 forks.
Language: Python
Topics: ai-agents, distillation, document-understanding, fine-tuning, grpo, knowledge-distillation, llm, multimodal, posttraining, qwen