OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
RAG & Knowledge Systems

Vectrain

Vectrain is a Go framework for building a pipeline that ingests data, generates vector embeddings and stores them in a vector database. Its README describes Kafka and REST sources, Ollama embedding and Qdrant storage, with a REST API for pipeline control.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Vectrain — OpenViST project profile in RAG & Knowledge Systems.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use Vectrain when you need a configurable ingestion-and-embedding backend that connects supported data sources to vector storage for a downstream application.

Who it is for

  • Developers building a semantic-search, recommendation or analytics backend around stored embeddings.
  • Teams processing data from Kafka topics or REST services into a Qdrant vector database.
  • Go developers extending a pipeline through custom source, embedder or storage components.

Documented uses

  • Ingest data from Kafka or a REST source.
  • Generate embeddings using the documented Ollama integration.
  • Store and query embeddings in Qdrant.
  • Control the pipeline through its HTTP API.
Before you choose

Requirements & limits

The repository is public and identifies the project as MIT-licensed. The README lists Go 1.24 or later and Docker for running dependencies such as Kafka, Ollama and Qdrant.

  • The documented source options are Kafka and REST.
  • The README names Ollama for embedding generation and Qdrant for vector storage.
  • YAML configuration selects pipeline components and allows batch-size and worker-count adjustments.
  • Custom source, embedder and storage types require implementing the corresponding Go interface and registering the component in the factory.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Install and configure Vectrain for a REST source

Prepare a local Vectrain installation and configure it to receive input through its REST source.

Before you start

  • Go 1.24 or later.
  • Docker for running dependencies such as Ollama and Qdrant.
  • A checkout of the Vectrain repository.

Steps

  1. Clone the repository and change into its directory: the command below followed by cd vectrain.

    Documented command
    git clone https://github.com/torys877/vectrain.git

  2. Install Go dependencies with the command below.

    Documented command
    go mod tidy

  3. Create config/config.yaml and adapt the REST example configuration in ./config/rest_config.yaml with the required endpoints and settings.

Expected outcome

The application dependencies are installed and a REST-oriented configuration file is prepared for running Vectrain.

Guide limitations

  • This guide covers setup and configuration, not starting the service or verifying a successful embedding run.
  • The README’s clone URL uses the torys877 owner, while the repository source URL identifies philpher0x; confirm the intended repository before cloning.
  • The README does not detail the configuration fields or provide specific endpoint values.
Documentation 1 ↗
Source-based practical guide

Submit text to Vectrain’s REST source and process it

Send a text-and-payload message to the REST source and have Vectrain process it through the configured embedding and storage stages.

Before you start

  • A running Vectrain application configured for REST input, an embedding service, and Qdrant storage.
  • The REST source is available at http://127.0.0.1/source/send; messages are embedded only after the pipeline is started.

Steps

  1. Start the application using the command below, or build it with the command below and run ./vectrain --config=config/config.yaml.

    Documented command
    go run cmd/vectrain/main.go --config=config/config.yaml
    Documented command
    go build -o vectrain cmd/vectrain/main.go

  2. Start the workflow by sending a POST request to /api/start.

  3. Send a message to http://127.0.0.1/source/send with a JSON body containing Text and Payload, for example the fields shown in the README.

  4. Allow the pipeline to send the source data to its embedding service and store generated embeddings in Qdrant; use POST /api/stop to stop the workflow when appropriate.

Expected outcome

Vectrain processes submitted messages through its source, embedder, and storage stages, storing generated embeddings in Qdrant.

Guide limitations

  • Ollama is identified as the currently supported embedding service and Qdrant as the currently supported vector database; other providers are mentioned in the workflow description but not established as supported configuration options.
  • The README gives endpoint paths and a sample message, but does not specify request headers, response formats, or a complete configuration.
  • The source endpoint can accept messages while the pipeline is stopped, but those messages are not embedded until the pipeline starts.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
17 October 2026
Original website title
philpher0x/vectrain · Official homepage title · retained in its source language
How we review sourcesSuggest a correction