Live Model Demo
Test the configured model endpoint before wiring it into workflows.
This page checks model URLs and runs a real prompt against OpenAI-compatible endpoints.
Connectivity
Endpoint settings
Connection path
Server
Model
llama3.1:8b
Endpoint
http://localhost:11434/v1
Health endpoint
http://localhost:11434/v1/models
Prompt limits
4/5 prompts remain, 500 input words, 1000 output tokens
Not reachable
No connection could be made because the target machine actively refused it. (localhost:11434)
Semantic Kernel Prompt
Run a live model response
The live model response will appear here.