marcusmi4n
/

phi-3.5-mini-instruct-quantized-qnn

8-bit precision

onnx_dynamic_int8

Model card Files Files and versions

phi-3.5-mini-instruct-quantized-qnn / model.model.layers.1.input_layernorm.weight

Commit History

Upload Phi-3.5 quantized for QNN deployment (50% compression, tested & verified)

597cb25
verified

marcusmi4n commited on Sep 9