Keep-Alive and Memory Control
50%
Forcing an Unload Right Now
You don't have to wait for the timer. Send a request with keep_alive: 0 and an empty payload to drop the model immediately:
curl http://localhost:11434/api/generate -d '{
"model": "granite3.3:2b",
"keep_alive": 0
}'
Or use the CLI shortcut:
Local AI Engineering with Ollama
Run, understand, customize, fine-tune, and build agentic apps on your own hardwareEnroll now to unlock all content and receive all future updates for free.