Shypd.aiThe largest open index of AI toolsDB-GPT-Hub is an open-source repository for models, datasets, and fine-tuning techniques to enhance Text-to-SQL parsing with Large Language Models. It supports Text2NLU and Text2GQL fine-tuning to improve semantic understanding and generate graph queries.
Trending
About
DB-GPT-Hub is an experimental open-source project focused on leveraging Large Language Models (LLMs) for Text-to-SQL parsing. It provides a comprehensive workflow including data collection, preprocessing, model selection, construction, and fine-tuning of model weights. The project aims to improve Text-to-SQL capabilities while reducing training costs, making it accessible for more developers to contribute. It supports various base models like CodeLlama, Baichuan2, LLaMa/LLaMa2, Falcon, Qwen, XVERSE, ChatGLM2, ChatGLM3, internlm, sqlcoder-7b, and sqlcoder2-15b. DB-GPT-Hub also supports Text2NLU fine-tuning for enhanced semantic understanding and Text2GQL fine-tuning for generating graph queries. The project has achieved high execution accuracy rates on the Spider dataset, demonstrating its effectiveness in enabling automated question-answering based on databases using natural language.
Capabilities
Pricing & Plans
Open Source
Free
FAQs
Trending
