简体中文 | English

Dash-CookBook is the repository for sharing various receipts to complete different tasks using DashScope APIs and other API services such as the vector-store DashVector.

DashScope aims to provide the efficient API services backed by the state-of-the-art models in different areas, including LLM, text-to-image, embedding, text-to-speech etc. Visit the official DashScope website for a complete list of models supported.

ModelStudio(百炼) is a one-stop large-scale model development platform based on DashScope API services. It provides complete model service tools and full-chain application development kits for enterprise customers and individual developers, pre-installing rich capability plugins, providing convenient integration methods such as RAG and Assistant API, and efficiently completing the construction of large-scale model applications.

DashVector is a fully-managed vectorDB service that supports high-dimension dense and sparse vectors, real-time insertion and filtered search. It is built to scale automatically and can adapt to different application requirements. It is currently integrated into 🦜️🔗LangChain as well.

This repository will host examples for accomplishing various tasks by leveraging the aforementioned API services. Most examples are presented via Colab notebook and are executable right out of the box (once you have the API-key for the underlying API services).

• Basic usage for LLM calls:

• Basic semantic search built with Text-Embedding and VectorDB APIs:

• Question-answering with LLM and knowledge-enhancement:

• Finetuning LLM to generate advertisement for e-commerce:

• Image Generation using LLM-enhanced prompt:

• Using API and SDK to call agent applications created on the ModelStudio platform:

• Building agents with Assistant API:

  • Assistant API Basic Usage:
  • Assistant API Best practice - Travel Assistant:
  • Assistant API Best practice - Official Documentation Assistant:
  • Assistant API Best practice - Multi-Agent:

• Building RAG applications:

  • RAG Application Quick Start:
  • Build rag application with llama index SDK:

• ... stay tuned for more examples, contribution also welcome!

If there are examples you'd like to see, feel free to open an issue. Quality examples are also welcome via pull request, please provide a ipynb implementation following existing examples.