---
title: docker ollama model
short_description: REST API to serve a model using Docker and Ollama
emoji: 🔥
colorFrom: green
colorTo: blue
sdk: docker
pinned: false
---

To do inference, use command line

curl https://myyim-docker-ollama-model.hf.space/api/generate -d '{
  "model": "gemma3:1b",
  "prompt": "Write me a poem about generative AI.",
  "stream": false
}'

(https://medium.com/p/1f5d8f871887)

or

import base64
import requests

image_path = 'your_image.jpg'
with open(image_path, 'rb') as image_file:
    image_data = image_file.read()
base64_encoded_data = base64.b64encode(image_data)
base64_string = base64_encoded_data.decode('utf-8')

headers = {
    'Content-Type': 'application/x-www-form-urlencoded',
}

model = "gemma3:4b"
prompt = "Describe the image in details and give it a caption."
ollamaURL = 'https://myyim-docker-ollama-model.hf.space/api/generate'

data = {
    "model": model, 
    "prompt": prompt, 
    "stream": False, 
    "images": [base64_string]
}

response = requests.post(ollamaURL, headers=headers, json=data)
print(response.json()["response"])

(https://medium.com/@manyi.yim/deploy-ollama-models-on-hugging-face-spaces-with-python-library-requests-503ac6b5ca04)