- INVENTORY
- SPECS
- DECISION
- HUMAN
Short answer
Can I manage my watsonx models using my AI?
Yes. Your AI uses this MCP to list available foundation models and fetch their technical specifications. You'll get a clear view of model details without leaving your chat interface.
Model management outcomes
What you can do with your AI
These are the specific ways your AI interacts with watsonx.
INVENTORY
List available models
Your AI pulls a list of foundation models from your watsonx instance. You'll see exactly what's available to use.
SPECS
Get technical details
The AI fetches specific parameters and technical specs for any model you select. This helps you understand model limits.
DECISION
Compare model options
By pulling specs for different models, your AI can help you pick the right one for your specific task.
HUMAN
Manual configuration
The AI only reads data. You still need to go to the watsonx console to change settings or deploy models.
The workflow
What your AI does when the request arrives.
The AI acts as a bridge to your watsonx model registry.
Scan registry
Your AI checks your watsonx environment to see which foundation models are currently active.
list_modelsFetch specs
The AI pulls the technical profile for a specific model you've identified.
get_model_detailsAnalyze data
Your agent uses the retrieved technical details to answer your questions about model capabilities.
get_model_detailsPresent info
The AI delivers the model list or specific technical details directly in your chat window.
list_models
Try it
Copy these to start.
Use these prompts to start talking to your watsonx models.
Starting points
These are opening lines. Swap out the model names for the ones you actually use in watsonx.
IBM watsonx Connector
You're all set. Choose your MCP client and follow the setup instructions.
https://edge.vinkius.com/vk_preview_gXytFT8fNy6ELfNWb5ZhGBzw4TcaHPZqjIMO9o5b/mcpClaude Desktop
Follow the steps below to connect in seconds.
- 1In Claude Desktop, open Settings → Connectors.
- 2Click “Add custom connector” and paste the connector link above as the remote MCP server URL.
- 3Click Add and start a new chat — IBM watsonx capabilities are ready to use.
{
"mcpServers": {
"ibm-watsonx-mcp": {
"url": "https://edge.vinkius.com/vk_preview_gXytFT8fNy6ELfNWb5ZhGBzw4TcaHPZqjIMO9o5b/mcp"
}
}
}List all the foundation models in my watsonx instance.
What are the technical specifications for the Granite-20b-instruct model?
Show me the details for all available models in my watsonx account.
Which models in my watsonx registry support specific parameter configurations?
Claude
ChatGPT
Cursor
VS Code
Windsurf
Claude Code
JetBrains
Cline
Start here
Connect IBM watsonx once, then ask.
One connection lets your AI read your model registry. Your login credentials stay encrypted on our side.
Connect IBM watsonx to your AIFAQ
How this task behaves
- 01
Can my AI deploy a model to watsonx?
No. This MCP only provides tools to list and inspect models. You must handle deployment through the watsonx interface.
- 02
Can the AI update model parameters?
No. The AI can only read technical specifications; it cannot modify any model settings or configurations.
- 03
Can I use this with Claude or Cursor?
Yes. This works with any MCP-compatible client.
- 04
Does the AI see my private training data?
No. The AI only accesses the model metadata and technical specs provided by the tools.
- 05
Can the AI delete models from my account?
No. There are no tools in this MCP that allow for the deletion or removal of models.
More questions about IBM watsonx? See everything the IBM watsonx Connector can do
Connect IBM watsonx to Claude, Cursor, ChatGPT & more
