docs/multimodal/gemma3.md

Gemma 3 vision

This is very experimental, only used for demo purpose.

Quick started

You can use pre-quantized model from ggml-org's Hugging Face account

# build
cmake -B build
cmake --build build --target llama-mtmd-cli

# alternatively, install from brew (MacOS)
brew install llama.cpp

# run it
llama-mtmd-cli -hf ggml-org/gemma-3-4b-it-GGUF
llama-mtmd-cli -hf ggml-org/gemma-3-12b-it-GGUF
llama-mtmd-cli -hf ggml-org/gemma-3-27b-it-GGUF

# note: 1B model does not support vision

How to get mmproj.gguf?

Simply to add --mmproj in when converting model via convert_hf_to_gguf.py:

cd gemma-3-4b-it
python ../llama.cpp/convert_hf_to_gguf.py --outfile model.gguf --outtype f16 --mmproj .
# output file: mmproj-model.gguf

How to run it?

What you need:

• The text model GGUF, can be converted using convert_hf_to_gguf.py
• The mmproj file from step above
• An image file

# build
cmake -B build
cmake --build build --target llama-mtmd-cli

# run it
./build/bin/llama-mtmd-cli -m {text_model}.gguf --mmproj mmproj.gguf --image your_image.jpg