Managing Models
AI models are managed through the Executors panel within each node's detail view. An executor is the process a Node runs to serve a particular model — for each enabled model, the Node keeps a dedicated executor running.
To reach the Executors panel, open the Nodes page, click the edit icon on a node, and look at the right-hand Executors panel.
Executor types
| Type | Purpose |
|---|---|
| Language Model | A chat or text generation model — used for conversations and API completions |
| Embedding Model | Converts text to vector embeddings — used by applications that need semantic search or similarity matching |
Executor status
Each executor displays its current status:
| Status | Meaning |
|---|---|
| Online | The model is loaded and ready to serve requests |
| Starting | The model is loading — requests cannot be served yet |
| Offline | The executor is stopped |
| Stopping | The executor is in the process of shutting down |
Starting and stopping executors
Use the start and stop buttons on each executor row to control whether it is running. Stopping an executor unloads the model from memory, which frees up RAM and VRAM. Starting it again reloads the model, which takes a few moments.
If a model shows Restart to apply in amber, its configuration has been updated and a restart is needed for the changes to take effect.
Installing new executors
Click the + button (top-right of the Executors panel) to open the executor installer. This lets you download and install AI model executors from the Parker Software executor feed.
To add GGUF model files manually instead, see Adding Models.
Editing executor parameters
Click the gear icon on an executor to open its parameter editor. This lets you adjust settings such as the number of GPU layers to offload, context length, and other inference parameters.
After saving parameters, the executor will show Restart to apply. Restart it to load the new configuration.
Refreshing executor discovery
Click the refresh button to prompt the Node to re-scan its model directory. Use this if you have just added a new GGUF file to the models folder and it has not appeared in the list yet.