RAG SDK

中文 | English

Quick Reference

• Where to get help
  • Issue Feedback
  • RAG SDK Code
  • RAG SDK API Reference
  • RAG SDK Documentation
  • Image Repository

RAG SDK

RAG SDK is knowledge enhancement development kit for large language models. It addresses the issues of slow knowledge updates and weak domain-specific knowledge answering in large models. It provides features such as domain-specific tuning, generation enhancement, and knowledge management for large model knowledge bases, helping users build exclusive, high-performance, and accurate large model question-answering systems.

Supported Tags and Dockerfile Links

Tag Naming Convention

Tags follow this pattern:

<ragsdk-version>-<chip-series>-<os>-<python-version>

Field	Example Values	Description
RAG SDK Version	26.0.0	RAG SDK version
Chip Series	910, A3, atlas 300I Pro	Target Atlas chip family
Operating System	ubuntu22.04, openeuler24.03	Base operating system
Python Version	py3.11	Python version

Version Notes

RAG SDK Image Matching Table

IMAGE version	RAG SDK version	CANN version
26.0.0	26.0.0	9.0.0

Quick Start

How to Build

# Clone the repository on the host, enter the docker directory, replace {cann version} with the actual version, and replace {your_repo} with the actual image repository
git clone https://gitcode.com/Ascend/RAGSDK.git && cd RAGSDK/docker

docker build --network host --build-arg CANN_VERSION={cann version} -t {your_repo}/ragsdk:latest -f Dockerfile.<chip-series>.<os> .

Run RAG SDK Container

 docker run -itd --name=rag_sdk_demo --network=host \
     --device=/dev/davinci_manager \
     --device=/dev/hisi_hdc \
     --device=/dev/devmm_svm \
     --device=/dev/davinci0 \
     -v /usr/local/Ascend/driver:/usr/local/Ascend/driver:ro \
     -v /usr/local/sbin:/usr/local/sbin:ro \
     -v /path/to/model:/path/to/model:ro \
     {image-name}:{image-tag} bash

Parameter Description

• /path/to/model: Model storage directory. Place model files in this directory if you need to load models.
• {image-name}:{image-tag}: Specify the RAG SDK image and tag to run.

Enter the Container

docker exec -it rag_sdk_demo bash

RAG SDK Usage

RAG SDK provides comprehensive sample code to help developers get started quickly. The sample code inside the container is located at /workspace/RAGSDK/example. You can also access the latest demo examples through the following link:

• RAG SDK Demo Samples

Development

# Use the RAG SDK image as the base image and add user software
FROM swr.cn-south-1.myhuaweicloud.com/ascendhub/ragsdk:26.0.0-910b-ubuntu22.04-py3.11
RUN apt update -y &&
    apt install gcc ...
...

Supported Hardware

Chip Series	Product Examples	Architecture
Atlas 910	Atlas 800T A2, Atlas 900 A2 PoD	ARM64/ X86_64
Atlas A3	Atlas 800T A3	ARM64/ X86_64
Atlas 300I Pro	Atlas 300I Pro、 Atlas 300V Pro	ARM64/ X86_64

License

View the license information for RAG SDK and Mind series software included in these images. As with all container images, pre-installed packages (Python, system libraries, etc.) may be subject to their own licenses.