5,382,281 AI resources indexedWorking toward the world’s largest AI resource index.

Browse resources 13 sections
PDF

Build document RAG system with Kimi-K2, Gemini embeddings and Qdrant

Knowledge & search workflow for n8n with 25 active steps. Starts with Manual Trigger, Chat Trigger, Execute Workflow Trigger. Uses Google Gemini. Integrations include Google Gemini, Qdrant.

By JimleukWorkflowReference information

What it does

25 active nodes · Starts with Manual Trigger, Chat Trigger, Execute Workflow Trigger, MCP Server Trigger.

Steps in the template include:

  • Kimi-K2 via Featherless.ai (featherless)
  • When clicking ‘Execute workflow’ (Manual Trigger)
  • Extract from File (Extract from File) — pdf
  • Get Response (Edit Fields (Set))

Nodes and integrations

  • featherless ×2
  • Manual Trigger
  • Extract from File
  • Edit Fields (Set) ×2
  • HTTP Request ×3
  • Split Out ×2
  • qdrant ×4
  • Loop Over Items (Split in Batches)
  • No Operation, do nothing
  • Chat Trigger
  • Execute Workflow Trigger
  • Google Gemini Chat Model
  • Call n8n Workflow Tool ×2
  • MCP Server Trigger
  • AI Agent
  • Execute Sub-workflow
See the author’s step names
  • Kimi-K2 via Featherless.ai (featherless)
  • When clicking ‘Execute workflow’ (Manual Trigger)
  • Extract from File (Extract from File) — pdf
  • Get Response (Edit Fields (Set))
  • Download Large Document (HTTP Request)
  • Split Pages (Split Out)
  • Retrieval Vectors with Gemini Embeddings 001 (HTTP Request)
  • Add Docs To Qdrant Vector Store (qdrant) — point / upsertPoints
  • Loop Over Items (Loop Over Items (Split in Batches))
  • Chunk Ref (No Operation, do nothing)
  • Create Collection (qdrant) — createCollection
  • Create Summary Index (qdrant) — payload / createPayloadIndex
  • When chat message received (Chat Trigger)
  • Subworkflow Trigger (Execute Workflow Trigger)
  • Google Gemini Chat Model1 (Google Gemini Chat Model)
  • Highway Code Manual1 (Call n8n Workflow Tool)
  • Highway Code Manual (Call n8n Workflow Tool)
  • Highway Code MCP Server (MCP Server Trigger)
  • Highway Code Expert (AI Agent)
  • Retrieval Vectors with Gemini-Embeddings-001 (HTTP Request)
  • Test Questions (Edit Fields (Set))
  • Split Out (Split Out)
  • Query Docs from Qdrant Vector Store (qdrant) — search / queryPoints
  • Call Highway Code Manual Tool (Execute Sub-workflow)
  • Kimi-K2 via Featherless ai (featherless)

Community node packages

Community nodes run third-party code and can access workflow data and the n8n host. Review the packages before running this template. Community node risks

Accounts and credentials

  • Google Gemini access
  • Qdrant

Credential types in the graph: featherlessApi, googlePalmApi, qdrantRestApi

Check the HTTP request steps for any additional service accounts or API keys.

AI models and swaps

Google Gemini · Model or API operation

Model ID not specified in the template in Retrieval Vectors with Gemini Embeddings 001. Provider-specific step. A different provider requires edits and testing.

Google Gemini · Chat

Model ID not specified in the template in Google Gemini Chat Model1. Potentially swappable with a compatible n8n chat model node. Match tool calling before testing.

Google Gemini · Model or API operation

Model ID not specified in the template in Retrieval Vectors with Gemini-Embeddings-001. Provider-specific step. A different provider requires edits and testing.

Resource information

Credit basis
Author listed by n8n
Pricing
Template access shown on n8n
License
Not specified
Last record update
Nov 5, 2026