4. Run First Inference

Once your model is deployed, you can run your first inference on the device. Inference means sending input data such as an image, sensor value, or text to the model and receiving results instantly. With Loc.ai, inference runs locally through Loc.ai:Link, delivering faster responses, offline capability, and improved data privacy.

Your first inference confirms that the model, device, and deployment setup are working successfully end-to-end.

Steps to Run Your First Inference

Step 1: From the left sidebar, open Inference, then click Run on your deployed model.

Step 2: View the inference results table to monitor requests, confidence scores, duration, and status for your model runs.

Step 3: Click Detailed Results to expand the view and see full inference information are:

• Model: The model used for inference
• Device: The device that ran the model
• Result: The classification or output
• Confidence: Confidence score as a percentage
• Duration: Time taken for the inference

Field	Description	Example
ID	Unique identifier assigned to the inference result.	eb9e2530-dea1-4cfc-a394-525ad99d0447
Model	Name of the model used for the inference request.	Qwen3-LLM
Device	Device or node where the inference was processed.	Welcome To Locai
Status	Current execution result of the inference task.	success
Type	Category of the deployed model.	language_models
Duration	Total time taken to complete the inference.	5.0 ms
Confidence	Confidence score returned by the model for the result.	100.0%
Start Time	Timestamp when the inference process started.	3/29/2026, 5:08:49 PM
End Time	Timestamp when the inference process finished.	3/29/2026, 5:08:54 PM
Created	Record creation date for the inference entry.	Invalid Date
Output	Displays the generated result. If the output is a .md file, the Markdown content should be rendered and shown directly in the Output section.	Rendered Markdown content

Now you can View and Export Results.