• VECTOR
  • INDEX
  • HUMAN
  • RAG

How to generate embeddings for RAG with your AI

Connect SambaNova to your AI client to turn text into vectors. Once you're set up, your agent handles the math required to build your retrieval index.

Ask AI about this page

Short answer

How do I get embeddings for RAG?

Your agent uses SambaNova to transform text into high-dimensional vectors. These representations allow you to populate a vector database for semantic search. You'll see the specific steps and prompts below.

The results of embedding

What you get from the process

The AI turns raw text into the mathematical data needed for retrieval.

VECTOR

High-dimensional vectors

The AI produces dense numerical representations of your text. These vectors capture the semantic meaning of your data.

INDEX

Populated vector stores

You get the data needed to fill your vector database. This enables your agent to perform similarity searches later.

HUMAN

Manual chunking decisions

You must decide how to split your documents before embedding. The AI follows your lead on how to structure the input.

RAG

Contextual retrieval

The resulting embeddings allow your agent to find relevant context during a query. This makes your RAG system much more accurate.

The workflow

What your AI does when text arrives.

The agent manages the conversion from text to vector data.

  1. Prepare text

    Your agent breaks your documents into specific chunks. This ensures the embeddings represent distinct pieces of information.

    create_embedding

  2. Run inference

    The agent sends the text chunks to SambaNova to generate the vector values.

    create_embedding

  3. Verify dimensions

    The AI checks that the generated embeddings match the expected high-dimensional format.

    create_embedding

  4. Ready for storage

    The agent provides the final vectors so you can move them into your database.

    create_embedding

Try it

Copy these to start.

Use these prompts to kick off the embedding process.

Starting points

Swap out the file names or text snippets with your own data.

SambaNova Connector

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

Connector linkhttps://edge.vinkius.com/vk_preview_z52tsMdkFaoCTCBeg6u7bbxNhhnjcT0r5KXU6hjj/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 — SambaNova capabilities are ready to use.
Configuration · claude_desktop_config.jsonCopy
{
  "mcpServers": {
    "sambanova-ai-inference-mcp": {
      "url": "https://edge.vinkius.com/vk_preview_z52tsMdkFaoCTCBeg6u7bbxNhhnjcT0r5KXU6hjj/mcp"
    }
  }
}
Copy into chat04

  • Generate embeddings for the contents of 'project_notes.txt'.

  • Create vectors for this text: [paste your text here].

  • Turn the data in 'user_manual.pdf' into embeddings for my RAG system.

  • Embed the contents of 'api_docs.md' so I can use it for retrieval.

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

Start here

Connect SambaNova once, then ask.

Link your account once and your credentials stay encrypted. You can find the full technical details on the Connector page.

Connect SambaNova to your AI

FAQ

How this task behaves

  • 01

    Can the AI automatically save these vectors to my database?

    No. The AI generates the embeddings, but you have to handle the actual storage in your vector database.

  • 02

    Does the AI change my original files?

    No. The tool only reads the text to create embeddings; it won't touch or modify your source files.

  • 03

    Can I embed images or audio?

    No. This tool is built specifically for text embeddings.

  • 04

    Will the AI decide how to split my documents?

    The AI can chunk the text, but you should tell it exactly how you want the data split for the best results.

  • 05

    Can the AI delete my vector database?

    No. This tool has no capability to delete or modify external databases.

  • More questions about SambaNova? See everything the SambaNova Connector can do

Connect SambaNova to Claude, Cursor, ChatGPT & more