Skip to content
FastVLM.net

FASTVLM GUIDE · Python

FastVLM Python Guide

Run image inference with Apple’s official predict.py entry point and a downloaded FastVLM checkpoint.

All FastVLM guides

Step by step

  1. 1

    Complete the repository setup.

  2. 2

    Download and extract an official checkpoint.

  3. 3

    Pass the checkpoint directory, image and prompt to predict.py.

  4. 4

    Treat model output as generated text and validate it for your use case.

Example

Python 
python predict.py \
  --model-path /path/to/checkpoint-dir \
  --image-file /path/to/image.png \
  --prompt "Describe the image."

Verify against the primary source

APIs, model files and dependency versions can change. Treat this page as an implementation map, then confirm the current command and license in the linked official documentation.

Open official source