Practical overview
When it fits
Use Qmedia when you want to build a queryable collection from creator materials and prefer a workflow that separates the web interface, retrieval service and multimodal model service.
Who it is for
- Creators who want to search and ask questions across a collection of text, image and short-video material.
- Teams or developers seeking a self-hosted content extraction and retrieval workspace.
Documented uses
- Extract and structure information from image, text and short-video materials.
- Search a media collection and get query-based answers grounded in its extracted content.
- Review source material and breakdowns through content cards.
- Use the model service independently for image or text encoding, video transcription extraction, or image OCR.
Requirements & limits
The project is open source under the MIT license. Its README describes local deployment and also says model services can be deployed in the cloud, depending on available resources.
- The combined setup consists of mm_server, mmrag_server and qmedia_web; the README also describes deploying these services separately.
- The documented web stack is TypeScript, Next.js, Tailwind CSS and shadcn/ui; the RAG and model services are described as Python applications using LlamaIndex.
- The README identifies Ollama language models, CLIP image embeddings, BGE text embeddings, OCR and Faster Whisper transcription among its model components.
- Several capabilities, including similar-media search and short-video highlight or style recognition, are marked as unchecked future items in the README.
Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.
Prepare custom media for Qmedia
Replace the sample content inputs used by the combined Qmedia workflow.
Before you start
- A Qmedia checkout with the combined services available.
- Image or video files for the assets/medias directory.
- Content-card data in assets/mm_pseudo_data.json.
Steps
Replace the existing files in assets/medias with your image or video files.
Replace the contents of assets/mm_pseudo_data.json with your content-card data.
Delete the historically stored db file.
Run the services so the model extracts information from the supplied material and stores it in the database.
Expected outcome
The services extract and store information from the replacement materials for retrieval and Q&A.
Guide limitations
- This guide covers custom-data preparation, not service installation or a full deployment procedure.
- The README describes image and video files in assets/medias; it does not specify other file types for that directory.
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 |
|---|---|---|---|
| Qmedia | Extract and structure information from image, text and short-video materials. | Self-hosted, Web | The project is open source under the MIT license. Its README describes local deployment and also says model services can be deployed in the cloud, depending on available resources. |
| 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
- 9 October 2026
- Original website title
- QmiAI/Qmedia · Official homepage title · retained in its source language