Automation library · AI assistants

Vision RAG and image embeddings with Cohere Command-A

Started by someone opening a chat with it, and works across Qdrant. It puts a language model behind a retrieval step, so answers are drawn from documents you supply rather than from the model alone. 27 steps, 1 branch, 1 loop or waits: large, and worth reading before you trust it with anything live.

LargeCalls a modelRetrieval

Download free
27 stepsLarge: read it before trusting it with anything live
1Branch: places the flow splits
1Loop or waits
1Service it connects to

The flow

Starts it 2Model 10Logic 9Service 6

The steps in the order they run, from what starts it.

  1. 1When clicking ‘Execute workflow’Manual triggerStarts it
  2. 2When chat message receivedChat triggerStarts it
  3. 3Simple MemoryMemory Buffer WindowModel
  4. 4Simple Memory1Memory Buffer WindowModel
  5. 5Technology Innovation Report ToolTool CodeModel
  6. 6Chat Model via Command-RLm Chat CohereModel
  7. 7Create CollectionQdrantService
  8. 8EmbeddingsEmbeddings CohereModel
  9. 9Technology and Innovation Report 2025SetLogic
  10. 10AI AgentAgentModel
  11. 11Respond to ChatChatModel
  12. 12Respond to Chat1ChatModel
  13. 13Get Relevant ImagesVector Store QdrantModel
  14. 14Split Out UrlsSplit OutLogic
  15. 15If has Tool Call?IfLogic
  16. 16Aggregate ResultsAggregateLogic
  17. 17Batch 5Split In BatchesLogic
  18. 18Get QuerySetLogic
  19. 19Image Understanding via Command-A-VisionHttp RequestService
  20. 20Page RefNo OpLogic
  21. 21Quick ConfirmationChatModel
  22. 22Download PageHttp RequestService
  23. 23Convert Image to Base64Extract From FileService
  24. 24Image Embeddings with Cohere Embed 4Http RequestService
  25. 25Prepare PointsSetLogic
  26. 26Aggregate PointsAggregateLogic
  27. 27Insert PointsQdrantService

How it's built

Starts on
Someone opening a chat with it
Steps
27, with 1 branch, 1 loop or waits
Size
Large: read it before trusting it with anything live
Connects to
Qdrant
Model
A language model, family not declared
Retrieval
Answers are drawn from documents you supply, not from the model alone

Questions

What does it need to run?

An n8n account, plus your own Qdrant account, and a key for a language model. You connect your own accounts after importing it; none are included.

What starts it?

It starts on someone opening a chat with it.

Does it use AI?

Yes. It calls a language model. Its answers draw on documents you give it, not on the model alone. Model calls are billed by the provider, not by n8n.

How long does it take to set up?

Take your time. At 27 steps it is large: read it before trusting it with anything live.

How do I import it?

Copy the workflow, open a new workflow in n8n and paste. The steps appear on the canvas, ready for your own accounts.

Before you run it

Nothing here has been tested by us. It is somebody's working automation, read and described from its own steps. Treat the connections it wants as a list of accounts it will touch, and read the 27 steps before you point it at anything that matters.