meta-llama

/

Llama-3.2-11B-Vision-Instruct

like 861

Follow

Meta Llama 7,359

Using the example python script (with python 3.12) leads to

python llama32vi.py 
  File "/Users/xxx/Documents/AI/mlx/scripts/vlm/llama32vi.py", line 6
    { "role": "user", "content": {"type":"image_url","image_url":{"url":"https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"}},{"type":"text","text":"Describe this image in one sentence."} }
                                                                                                                                                                                                                                   ^
SyntaxError: ':' expected after dictionary key

Is the example script wrong, or should I be configuring something differently?