Quick Start
Here is a documentation page that shows how to setup a same tool like https://tidb.ai (opens in a new tab) from deployment to usage.
Step 1: Deployment
As this tool is open source, you can deploy it on your own server. We provide a guide to deploy the entire RAG using Docker Compose.
Step 2: Configure
After deployment, you need to configure the tool to use your own settings.
Configure the LLM - Large Language Model
The LLM is used for extracting knowledge from docs and generating responses. You can change the default LLM to another one.
Configure the Embedding Model
The Embedding Model is a machine learning model that is trained to generate embeddings for a given input. We must translate text to vectors with this model before insert vector to database.
Configure the Reranker [Optional]
The Reranker is a machine learning model that is trained to rerank the results retrieved from the database. We use this model to improve performance of answers.
Note: Reranker is used to improve the performance of the answers. If you don't have a reranker model, you can skip this step. But we recommend you to use a reranker model to improve the performance of the answers.
Step 3: Add Data Source
After deployment and configuration, you need to add data source to the tool for example your own docs, knowledge base, etc.
After adding data source, there will be a period of time for the tool to index the data. During this time, the tool will not be able to provide answers to users.
Step 4: Usage
After deployment, configuration and adding data source, you can use the tool to chat with users to answer their questions.
pingcap/tidb.ai provides several features to help you chat with users:
- Out-of-the-box chat interface, e.g. https://tidb.ai (opens in a new tab)
- API to chat with users programmatically, e.g. https://tidb.ai/api-docs (opens in a new tab)
- Embeddable chat widget to integrate with your website