madnesslab
/

Qwen2-VL-2B-Instruct-OpenVINO-INT4-ASYM-32-0.5

Model card Files Files and versions Community

Original model: https://huggingface.co/Qwen/Qwen2-VL-2B-Instruct

Quantitation documentation: https://docs.openvino.ai/nightly/notebooks/qwen2-vl-with-output.html

Quantitation config:

import nncf

compression_configuration = {
    "mode": nncf.CompressWeightsMode.INT4_ASYM,
    "group_size": 32,
    "ratio": 0.5,
}

Downloads last month: 1

Inference Providers NEW

This model is not currently available via any of the supported third-party Inference Providers, and HF Inference API was unable to determine this model's library.