keisuke-miyako/gemma-2-2B-it-onnx-int4

optimum-cli export onnx \
  --model google/gemma-2b-it \
  --opset 17 \
  --framework pt \
  ~/.onnx/gemma-2b-it-pt

c.f. https://github.com/huggingface/optimum/issues/1755

fail

python -m onnxruntime_genai.models.builder \
  -m ~/.onnx/gemma-2b-it-pt \
  -o ~/.onnx/gemma-2b-it-onnx-int4-cpu \
  -p int4 \
  -e cpu

python \
  -m onnxruntime_genai.models.builder \
  -m google/gemma-2-2b-it \
  -p fp16 \
  -e cpu \
  -o ~/.onnx/gemma-2b-it-onnx-fp16-cpu