• VECTOR
  • DATABASE
  • SEARCH
  • HUMAN

How to generate text embeddings with your AI

Connect your AI client to NVIDIA's API to turn text into math. Once linked, your agent handles the vectorization for you.

Ask AI about this page

Short answer

How do I turn text into embeddings?

Your agent sends text to the NVIDIA API to get high-dimensional numerical arrays. These vectors represent semantic meaning, which you can then use for similarity searches or to populate a vector database.

The results

Where your data lands

The process converts raw text into usable mathematical representations.

VECTOR

Numerical arrays

The AI produces dense vectors that capture the semantic context of your input text.

DATABASE

Vector storage

You get the arrays needed to fill up vector databases for long-term storage.

SEARCH

Semantic retrieval

These embeddings let your agent find information based on meaning instead of just matching keywords.

HUMAN

Manual selection

You decide which specific text chunks are worth vectorizing and how to organize your final index.

The workflow

What your AI does when text arrives.

The AI acts as the bridge between your raw text and the math required for semantic search.

  1. Text ingestion

    Your agent reads the raw text from the documents or logs you want to process.

    nvidia_generate_embeddings

  2. Vectorization

    The AI converts those text segments into precise numerical arrays via the NVIDIA API.

    nvidia_generate_embeddings

  3. Data formatting

    The agent organizes the resulting vectors into a format your database can accept.

    nvidia_generate_embeddings

  4. Dimension check

    The AI verifies the output dimensions to ensure they match your target vector store.

    nvidia_generate_embeddings

Try it

Copy these to start.

Use these prompts to begin the embedding process.

Starting points

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

NVIDIA API Catalog Connector

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

Connector linkhttps://edge.vinkius.com/vk_preview_LgfbZayzkE3ZBfq9HWv3rgeK69AXP0h3anUpakUy/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 — NVIDIA API Catalog capabilities are ready to use.
Configuration · claude_desktop_config.jsonCopy
{
  "mcpServers": {
    "nvidia-api-catalog-mcp": {
      "url": "https://edge.vinkius.com/vk_preview_LgfbZayzkE3ZBfq9HWv3rgeK69AXP0h3anUpakUy/mcp"
    }
  }
}
Copy into chat04

  • Generate embeddings for the text in [document.txt] and prepare them for my vector store.

  • Convert the latest entries in [database_table] into numerical vectors.

  • Take these support logs from [folder_path] and create embeddings for semantic search.

  • Embed the content of [file_name] so I can perform similarity lookups.

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

Start here

Connect NVIDIA API Catalog once, then ask.

Link your account once. Your credentials stay encrypted on our side, and you can immediately start asking your agent to vectorize data.

Connect NVIDIA API Catalog to your AI

FAQ

How this task behaves

  • 01

    Can the AI automatically update my vector database?

    No. The AI generates the embeddings, but you must use your own tools to write those vectors into your database.

  • 02

    Does the AI store my raw text?

    No. The AI sends the text to the API to get the math, but it doesn't keep a permanent copy of your data.

  • 03

    Can I choose the embedding model?

    The AI uses the specific models provided through the NVIDIA API Catalog tools.

  • 04

    Will the AI delete my original files?

    No. The AI only reads the text to create embeddings; it has no capability to delete or modify your source files.

  • 05

    Can it embed images?

    No. This tool is specifically for converting text into numerical arrays.

  • More questions about NVIDIA API Catalog? The Connector page answers them. See everything the NVIDIA API Catalog Connector can do

Connect NVIDIA API Catalog to Claude, Cursor, ChatGPT & more