Practical overview
When it fits
Use it when you want to study or assemble the stages of a document-based RAG application in Python and can manage model configuration, dependencies and any local model deployment yourself.
Who it is for
- Developers learning how document ingestion, embeddings, vector storage and chat fit together in a RAG application.
- Teams prototyping a self-hosted or server-based document question-answering workflow with Qdrant.
- Python users who want to compare OpenAI and word2vec embeddings or OpenAI and ChatGLM chat configurations.
Documented uses
- Prepare documents in a corpus and convert them into vectors.
- Create a persistent Qdrant collection populated from the corpus.
- Run the separate knowledge-base application against a configured chat-model option.
- Follow the included materials to learn about word2vec training and building a knowledge base from scratch.
Requirements & limits
The project is published as a public GitHub repository and identified as open source. Its README describes running the components from a local Python environment.
- The repository targets Chinese dataset processing and describes a Chinese-language RAG workflow.
- Knowledge-base generation and the application are separate components with separate requirements files.
- OpenAI settings require an API key; word2vec model files must be downloaded separately.
- The local ChatGLM2-6B option requires a separate model deployment and port configuration.
Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.
Create a persistent Qdrant knowledge base
Vectorize documents in the corpus folder and store them in a named Qdrant collection.
Before you start
- A Python environment with the dependencies listed in vector_db/requirements.txt.
- Documents placed in the corpus folder.
- Project configuration for paths and COLLECTION_NAME.
- If using OpenAI embeddings, an OpenAI API key; alternatively, configure word2vec and obtain its model data.
Steps
Configure the embedding model type and relevant project paths in config.json.
Change to the vector_db directory.
Install dependencies with pip install -r requirements.txt.
Run python main.py to create the configured collection and store vectorized corpus documents.
Expected outcome
A persistent Qdrant collection populated with vectorized documents from corpus.
Guide limitations
- The README says word2vec model files are large and must be downloaded separately.
- The steps cover knowledge-base creation, not running or deploying the application.
Compare options for Retrieval-augmented applications
These projects document a common task and share a project role. This is a source-based comparison, not an execution benchmark or a guarantee of interchangeability.
| Project | Documented purpose | Platform | Access and requirements |
|---|---|---|---|
| LLM Corpus | Prepare documents in a corpus and convert them into vectors. | Self-hosted, Server | The project is published as a public GitHub repository and identified as open source. Its README describes running the components from a local Python environment. |
| Rag Powered AI Assistant Transforming Data Retrieval And Analysis Across The Web And Pdfs | Retrieve relevant passages from web and PDF content to answer a question. | Web | |
| LangChain in Supabase Edge Functions | Initialize and start a Supabase project locally. | Cloud, Server |
Sources & review
Source record details
- Source availability
- Active
- Sources checked
- 18 October 2026
- Original website title
- OceanPresentChao/llm-corpus · Official homepage title · retained in its source language