• SEARCH
  • GROUPING
  • RECO
  • HUMAN

How to generate vector embeddings with your AI

Connect IBM watsonx to your AI client to turn text into mathematical vectors. This lets your agent understand meaning instead of just matching keywords.

Ask AI about this page

Short answer

How can I create text embeddings with my AI?

You use the generate_embeddings tool to convert text into high-dimensional vectors. These vectors represent semantic meaning, which lets you build search or grouping workflows. Here is how the process works.

Where your data goes

The outcomes of embedding text

Once your agent generates these vectors, you can use them for several specific tasks.

SEARCH

Semantic search

The AI creates vectors for your documents. You use these to find content based on intent rather than exact keyword matches.

GROUPING

Automated clustering

The agent turns large datasets into vectors. You can then group similar topics or feedback together automatically.

RECO

Recommendation engines

Your AI generates embeddings for items. This allows you to suggest related content based on mathematical similarity.

HUMAN

Manual review

The AI provides the raw vector data. You decide which specific clusters or search results are accurate enough to use.

The workflow

What your AI does when the text arrives.

The AI handles the mathematical conversion so you can focus on your application logic.

  1. Input preparation

    Your agent takes the text or document snippets you provide and prepares them for the model.

    generate_embeddings

  2. Vector conversion

    The tool transforms the text into a list of numbers that represent the semantic context.

    generate_embeddings

  3. Data structuring

    The AI organizes these vectors into a format ready for your vector database.

    generate_embeddings

  4. Result delivery

    The agent returns the completed embeddings to your chat interface or development environment.

    generate_embeddings

Try it

Copy these to start.

Use these prompts to trigger the embedding process.

Starting points

Swap the bracketed text for your actual content or file names.

IBM watsonx Connector

You're all set. Choose your MCP client and follow the setup instructions.

Connector linkhttps://edge.vinkius.com/vk_preview_gXytFT8fNy6ELfNWb5ZhGBzw4TcaHPZqjIMO9o5b/mcp

Claude Desktop

Follow the steps below to connect in seconds.

  1. 1In Claude Desktop, open Settings → Connectors.
  2. 2Click “Add custom connector” and paste the connector link above as the remote MCP server URL.
  3. 3Click Add and start a new chat — IBM watsonx capabilities are ready to use.
Configuration · claude_desktop_config.jsonCopy
{
  "mcpServers": {
    "ibm-watsonx-mcp": {
      "url": "https://edge.vinkius.com/vk_preview_gXytFT8fNy6ELfNWb5ZhGBzw4TcaHPZqjIMO9o5b/mcp"
    }
  }
}
Copy into chat04

  • Generate embeddings for this paragraph: [insert text here]

  • Create vectors for the contents of my customer_feedback.csv file.

  • Convert these product descriptions into embeddings: [insert list]

  • Use the generate_embeddings tool on the text in my current project file.

  • Claude
  • ChatGPT
  • Cursor
  • VS Code
  • Windsurf
  • Claude Code
  • JetBrains
  • Cline

Start here

Connect IBM watsonx once, then ask.

Link your account once. Your credentials stay encrypted, and you can start generating embeddings in any compatible client.

Connect IBM watsonx to your AI

FAQ

How this task behaves

  • 01

    Can the AI store these embeddings for me?

    No. The AI generates the vectors, but you must save them in your own vector database.

  • 02

    Does this work for image embeddings?

    No. This tool is specifically for text-based embeddings.

  • 03

    Can the AI delete my existing vectors?

    No. The tool only handles the generation of new embeddings.

  • 04

    How do I use these for semantic search?

    You take the generated vectors and compare them against your document vectors using a similarity metric.

  • 05

    Do I need to write code to trigger this?

    No. If you use an MCP-compatible client like Claude or Cursor, you can just ask your agent to do it.

  • More questions about IBM watsonx? The Connector page answers them. See everything the IBM watsonx Connector can do

Connect IBM watsonx to Claude, Cursor, ChatGPT & more