Connect an AI model for analysis

ParcelType runs on Windows 10 and 11. You choose the model that fits your privacy needs. The Test button shows response time and context window details for your selected endpoint.

Choose your model source

Cloud provider

Use a standard API endpoint to connect to a cloud AI provider. This option sends your prompts to an external server. The agent reads your logs and sends them for analysis. You control which provider you select. The connection requires an internet link. Data leaves your machine during processing.

Local model

Run a model directly on your Windows PC. This keeps all data inside your hardware. No prompts leave your machine during analysis. You manage the model resources yourself. This suits sensitive lab environments. You keep full control over your secrets.

Built-in model

Use the model included with ParcelType. This requires no extra setup or keys; it is ready to use without your own endpoint. You can switch to other models later.

Verify your connection

Press the Test button to check your endpoint. The tool shows the response time clearly. It also displays the context window size. You see if the model is reachable. This helps you avoid errors during runs.

Select the right model size

Use a bigger model for complex tasks, such as summarising long connection logs or extracting IOCs from behaviour logs. Send a hard step to a bigger model with run_on_model.

Model quality varies between different providers. Always review the diff before accepting changes. Use Ask first to pause before actions. This lets you check the output carefully. You decide if the change is correct.