Base on https://huggingface.co/Qwen/Qwen2-0.5B-Instruct
Convert to onnx model using https://github.com/microsoft/onnxruntime-genai
Using command: python src/python/py/models/builder.py -m Qwen/Qwen2-0.5B-Instruct -o path-to-onnx-model -p int4 -e dml
Inference Providers
NEW
This model isn't deployed by any Inference Provider.
๐
Ask for provider support
HF Inference deployability: The model has no library tag.