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

Pomaidb

PomaiDB is a C++ embedded vector database and multimodal memory engine for edge applications, providing local similarity search, offline RAG, and storage for data such as graph relationships, time series, and key-value records.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Pomaidb — 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 PomaiDB when an application needs its own local database for semantic retrieval or RAG, particularly on an edge device or in an offline setting.

Who it is for

  • Developers embedding local vector retrieval in edge software
  • Teams building offline RAG or on-device AI memory
  • IoT and edge application developers working with ARM64 devices

Documented uses

  • Store vectors and retrieve similar items locally.
  • Ingest documents and retrieve context for a query through the RAG pipeline.
  • Keep vector, graph, text, time-series, and key-value data within one embedded database.
  • Link vector search results to related graph or mesh objects.
Before you choose

Requirements & limits

The project is published as an open-source repository. Repository metadata lists an Apache-2.0 license; a README badge displays MIT, so check the repository licensing materials before adopting it.

  • The documented design is embedded and single-process, rather than a distributed database deployment.
  • The RAG example requires an embedding provider; its sample uses a mock provider.
  • C++ source builds require C++20, CMake 3.20 or later, and Git; Vulkan headers are needed for mesh-related functionality.
  • The README includes C++ and Python quick-start examples.
  • License signals differ between repository metadata and the README badge.

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

Build PomaiDB from source

Configure and compile the repository as a Release build.

Before you start

  • C++20 compiler
  • CMake 3.20 or later
  • Git
  • Vulkan headers for mesh-related functionality, located at third_party/vulkan/include/

Steps

  1. Clone the repository with its submodules using git clone --recursive https://github.com/pomagrenate/pomaidb.

  2. Create a build directory and enter it.

  3. Configure with cmake .. -DCMAKE_BUILD_TYPE=Release.

  4. Compile with make -j$(nproc).

Expected outcome

A Release build produced from the source checkout.

Guide limitations

  • Vulkan headers are specifically needed for mesh-related functionality.
  • The guide covers building only; it does not verify the build or describe deployment.
Documentation 1 ↗
Source-based practical guide

Ingest a document and retrieve context with the C++ RAG example

Use the documented C++ quick-start pattern to ingest a text document and request retrieved context.

Before you start

  • PomaiDB source or library available to the C++ application
  • C++20 compiler
  • An embedding provider; the example uses MockEmbeddingProvider

Steps

  1. Open a database with a path and vector dimension, as shown in the example.

  2. Create a membrane with kind kRag.

  3. Construct a RagPipeline with the database, membrane name, dimension, and embedding provider.

  4. Call IngestDocument with a document ID and text, then RetrieveContext with a query and result count.

Expected outcome

The example requests retrieved context for a query after document ingestion.

Guide limitations

  • The example uses a mock embedding provider and does not document integration with a particular production model.
  • The guide describes the example’s code stages, not a tested execution result.
Documentation 1 ↗
COMPARE OPTIONS

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.

ProjectDocumented purposePlatformAccess and requirements
PomaidbStore vectors and retrieve similar items locally.Linux, Self-hosted, ServerThe project is published as an open-source repository. Repository metadata lists an Apache-2.0 license; a README badge displays MIT, so check the repository licensing materials before adopting it.
LangsearchCrawl website or local data for a knowledge base.CLI, Self-hosted, Server
Hybrid Search RagCombine document retrieval and language-model generation in a configurable workflow.
Evidence and transparency

Sources & review

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