Ollama Vision Example Doesn't Send Image To Model #1809
Description
Describe the issue as clearly as possible:
When following the ollama vision example, the image doesn't seem to be passed along with the request to the model.
Steps/code to reproduce the bug:
# Just run the example in the documentation as is
import io
import requests
import PIL
import ollama
import outlines
from outlines.inputs import Image
# Create the model
model = outlines.from_ollama(
ollama.Client(),
"qwen2.5vl:3b"
)
# Function to get an image
def get_image(url):
r = requests.get(url)
return PIL.Image.open(io.BytesIO(r.content))
# Create the prompt
prompt = [
"Describe the image",
Image(get_image("https://picsum.photos/id/237/400/300"))
]
# Generate text
response = model(prompt)
print(response) # The image shows a black puppy with a curious and attentive expression.Expected result:
A description of the imageError message:
The model states that an image wasn't provided in various ways depending on the model chosen, but it doesn't work for any vision model. Example output:
I don't have the ability to view or process images. My capabilities are limited to **text-based interactions** only.
If you'd like me to describe an image, **please provide a detailed text description of what you see** (e.g., "a red apple on a white plate," "a city skyline at sunset with tall skyscrapers," etc.). I’ll then analyze the description and give you a thoughtful response.
If you accidentally tried to attach an image, note that I **cannot see or interpret visual content**—I can only work with the text you type. 😅 Let me know how I can help!Outlines/Python version information:
pyproject.toml:
[project]
name = "blah"
version = "0.1.0"
description = "Add your description here"
readme = "README.md"
requires-python = ">=3.10"
dependencies = [
"outlines[ollama]>=1.2.9",
"requests>=2.32.5",
]
uv sync
Context for the issue:
I'm trying to learn and start using Outlines, but haven't been able to get this to work.