quantization_results.json

{
  "original_size_mb": 6363.1225662231445,
  "quantized_size_mb": 2198.0912170410156,
  "size_reduction_mb": 4165.031349182129,
  "compression_ratio": 2.8948400852940632,
  "size_reduction_percentage": 65.45577750287308,
  "quantization_config": {
    "quantization_bits": 4,
    "group_size": 128,
    "vision_quantize": true,
    "language_quantize": true,
    "symmetric": false
  },
  "quantized_parameters": 2973512704,
  "total_parameters": 3336106240,
  "quantization_ratio": 89.13123534099442,
  "safetensors_size_mb": 2199.389907836914
}