ollama + fastAPI server
10K+
ollama/ollama, and was developed to allow external systems to release GPU memory occupied by Ollama.docker run -itd --gpus=all -v F:\docker_volume\llama:/root/.ollama -p 11434:11434 -p 5000:5000 --name ollama shinejh0528/ollama:1.0.0
curl -X POST http://localhost:5000/stop -H "Content-Type: application/json" -d '{"model":"gemma3:latest"}'
{"message":"Model 'gemma3:latest' stopped successfully.","output":""}
Content type
Image
Digest
sha256:cf36f61cd…
Size
3.1 GB
Last updated
8 months ago
docker pull shinejh0528/ollama:1.1.0