quantization_report.json

{
  "block_size": 128,
  "graph_contract_verified": true,
  "method": "weight-only-int4",
  "onnxruntime_version": "1.28.0",
  "op_types": [
    "MatMul",
    "Gather"
  ],
  "per_channel": null,
  "quantized_model_bytes": 123012561,
  "quantized_model_sha256": "4e4a5b343832c97946800acaa1dc48b81df13c5889f4b6d915cfbd7cfcf6944f",
  "semantic_parity_verified": false,
  "size_ratio": 0.13370504387717216,
  "source_model_bytes": 920029323,
  "source_model_sha256": "35bb5c1821d3dfa143c760a78699d4d9c4e48e9299134dadac10c7590aefd168",
  "symmetric": true,
  "web_runtime_verified": false,
  "weight_type": "int4"
}