{
  "axquant_version": "1.0.0",
  "calibration": null,
  "created_at": "2026-08-02T18:28:14.847095Z",
  "effective_bpw": 7.379967222761561,
  "files": [
    {
      "path": "README.md",
      "sha256": "223a9bd9a8922e7700cb3ae7e8710836106744cdc7650c8acef38f2381b4e966",
      "size_bytes": 7340
    },
    {
      "path": "axquant_plan.json",
      "sha256": "23e80e5d2d54a09be0360f87303e1aed3aded6f79fc629e746f6aff5c99808a1",
      "size_bytes": 203417
    },
    {
      "path": "axquant_quantizer_execution.json",
      "sha256": "3429b9c73510d99880237a90ed3e88c5e10faa08d6f2b265e94e14cde9384ed6",
      "size_bytes": 41265
    },
    {
      "path": "axquant_runtime.json",
      "sha256": "44f98fdd9f254275dbf0ae2b5d6a83cc97fc96958fb5983633782c3427da7401",
      "size_bytes": 1643
    },
    {
      "path": "chat_template.jinja",
      "sha256": "7451a05cf1e28a79d97d7c0bc951028c0b1915119bf9046acd06a0e3d931f47c",
      "size_bytes": 9062
    },
    {
      "path": "config.json",
      "sha256": "e3debd12d4d8eceeaa4ac74aae8930b86ec1b7c61f52264a36dd045f208edd89",
      "size_bytes": 47172
    },
    {
      "path": "generation_config.json",
      "sha256": "92afd6424501426eddcf7e1542f013d19e5987544977b4ee7bd26359bd5fd2ab",
      "size_bytes": 213
    },
    {
      "path": "model-manifest.json",
      "sha256": "7e51ebe6c2ce5c2276105322a5026eb653542f7b3b7da30aab221bd8b43c0b01",
      "size_bytes": 86992
    },
    {
      "path": "model.safetensors",
      "sha256": "548256a68c54ccdacc31faa581d2929f7233a2ac50f22074a40b98c2f05654fb",
      "size_bytes": 996941625
    },
    {
      "path": "model.safetensors.index.json",
      "sha256": "c147a4dff82e95025c8885867a36a676f0a74e6dfc82150f084f07fffc2972be",
      "size_bytes": 39286
    },
    {
      "path": "tokenizer.json",
      "sha256": "3e065a558a034185fe299917b398685c1facd0169a9eea1e629eb30c171fed81",
      "size_bytes": 9894271
    },
    {
      "path": "tokenizer_config.json",
      "sha256": "b99b837295c979cc956b9e018fe710123a5547df9d4143894ba7b95a7f08c42b",
      "size_bytes": 460
    }
  ],
  "format": "mlx",
  "logical_parameters": 1080632832,
  "main_logical_parameters": 1080632832,
  "main_weight_file_size_bytes": 996941625,
  "measured_main_bpw": 7.380428174886324,
  "measured_total_bpw": 7.380428174886324,
  "mtp_acceptance_retention": null,
  "mtp_distribution": {},
  "mtp_measured_speedup": null,
  "mtp_policy": {
    "candidate_bits": [
      8,
      16
    ],
    "min_bits": 8,
    "mode": "protected",
    "optimize_for_acceptance": true,
    "preserve_external_sidecar": true,
    "protect_norms": true,
    "protect_output_head": true
  },
  "mtp_present": false,
  "mtp_weight_file_size_bytes": 0,
  "plan_sha256": "14792f75cdcd65eacae9bc6493b1baf59ebb0358d05e774bd99ac12dca72ad87",
  "profile": "general",
  "protected_weight_file_size_bytes": 0,
  "quantizer": "axquant",
  "runtime": {
    "ax_engine": {
      "decode_kernel": null,
      "fused_mtp": null,
      "kernel_evidence": "unmeasured",
      "model_manifest": "model-manifest.json",
      "preferred_group_size": 32
    },
    "compatible_runtimes": [
      {
        "compatibility_level": "B",
        "manifest": "config.json",
        "mtp_support": "runtime-dependent",
        "name": "mlx-lm",
        "notes": [
          "Standard backbone inference is the compatibility target.",
          "AXQuant MTP metadata may be ignored by MLX-LM."
        ],
        "standard_inference": true,
        "standard_mlx_weights": true,
        "support_level": "standard-inference"
      }
    ],
    "created_at": "2026-08-02T18:28:14.372317Z",
    "kv_cache": null,
    "memory_policy": {
      "kv_cache_precision": "runtime-default",
      "mtp_buffers": "not-required",
      "prefix_cache": "runtime-managed",
      "unified_memory_safety_margin": "benchmark-required"
    },
    "mtp": {
      "acceptance_retention": null,
      "detected": false,
      "draft_tokens": null,
      "enabled_by_default": false,
      "head_precision": null,
      "measured_speedup": null,
      "optimized": false,
      "recommended_temperature_max": null,
      "sidecar_file": null,
      "verification_mode": null
    },
    "optimization_scope": "text-path",
    "primary_runtime": {
      "compatibility_level": "A",
      "manifest": "model-manifest.json",
      "mtp_support": "native",
      "name": "ax-engine",
      "notes": [
        "Runtime claims require a passing AX Engine doctor and benchmark report."
      ],
      "standard_inference": true,
      "standard_mlx_weights": true,
      "support_level": "optimized"
    },
    "schema_version": "axquant.runtime.v1"
  },
  "schema_version": "axquant.artifact.v2",
  "software_versions": {
    "ax_engine": null,
    "axquant": "1.0.0",
    "mlx": "0.32.0",
    "mlx_lm": "0.31.3",
    "pydantic": "2.13.4",
    "python": "3.13.14",
    "safetensors": "0.8.0"
  },
  "source_model": {
    "architecture": "LlamaForCausalLM",
    "format": "mlx",
    "local_path": null,
    "model_id": "openbmb/MiniCPM5-1B",
    "revision": null
  },
  "target_class": "4bit",
  "weight_distribution": {
    "4bit": {
      "fraction": 0.6287771645272388,
      "parameters": 679477248
    },
    "8bit": {
      "fraction": 0.18557659369727533,
      "parameters": 200540160
    },
    "bf16": {
      "fraction": 0.18564624177548586,
      "parameters": 200615424
    }
  },
  "weight_file_size_bytes": 996941625
}