{
  "schemaVersion": 1,
  "identity": {
    "modelId": "katara/llama-3.1-8b-instruct@1",
    "family": "llama-3.1",
    "bundleVersion": 1
  },
  "source": {
    "sourceWeightsHash": "0x0000000000000000000000000000000000000000000000000000000000000000",
    "tokenizerId": "meta-llama/Llama-3.1-8B-Instruct",
    "tokenizerHash": "0x46f4692df2d1efb5b9cccd3b30f01b45f8185eb60c233b1c1913bc0200e5f03a",
    "chatTemplateHash": "0xee216d5feb7a19e47433bf5de0c6b7131e4430566bb3646c3a66558359b47d66",
    "contextLimit": 131072,
    "license": "llama3.1-community"
  },
  "capabilities": {
    "workload": "chat",
    "tools": false,
    "structuredOutputs": false,
    "reasoning": false,
    "streaming": true,
    "maxOutputTokens": 8192
  },
  "inference": {
    "defaultParams": {
      "temperatureCapMilli": 1000,
      "stopPolicy": "standard"
    }
  },
  "validation": {
    "goldenSetId": "golden:llama31-8b:v1",
    "qualityEvalId": "eval:general-chat:v1",
    "minQualityPermille": 900,
    "maxTtftMs": 15000,
    "minTokensPerSec": 20,
    "minSuccessPermille": 990,
    "probeCadenceSec": 600
  },
  "profiles": [
    {
      "runtimeFamily": "mlx-metal",
      "runtimeVersion": "mlx-lm-0.31.3",
      "packaging": {
        "type": "native",
        "runtimeManifest": {
          "mlxLmVersion": "0.31.3",
          "pythonLockHash": "0xddda3a2db1b9f173ef5e71e192099f96abbdb2bd7c8726a46f6c0a1bab05e19d",
          "pythonRequirements": [
            "Jinja2==3.1.6",
            "MarkupSafe==3.0.3",
            "PyYAML==6.0.3",
            "Pygments==2.21.0",
            "annotated-doc==0.0.5",
            "anyio==4.15.1",
            "certifi==2026.7.22",
            "click==8.5.0",
            "filelock==3.32.6",
            "fsspec==2026.7.0",
            "h11==0.16.0",
            "hf-xet==1.6.0",
            "httpcore==1.0.9",
            "httpx==0.28.1",
            "huggingface_hub==1.30.0",
            "idna==3.19",
            "markdown-it-py==4.2.0",
            "mdurl==0.1.2",
            "mlx-lm==0.31.3",
            "mlx-metal==0.32.2",
            "mlx==0.32.2",
            "numpy==2.5.3",
            "packaging==26.3",
            "protobuf==7.36.1",
            "regex==2026.9.3",
            "rich==15.0.0",
            "safetensors==0.8.0",
            "sentencepiece==0.2.2",
            "shellingham==1.5.4",
            "tokenizers==0.23.2",
            "tqdm==4.70.0",
            "transformers==5.17.0",
            "typer==0.27.2",
            "typing_extensions==4.16.0"
          ]
        }
      },
      "artifacts": {
        "weightsHash": "0xc40ca166dff6083ef32edff4f41c5bdff889934dc32efb898cb24fc568f0ba0c",
        "source": {
          "hfRepo": "mlx-community/Meta-Llama-3.1-8B-Instruct-4bit",
          "hfRevision": "241a666dad6cb93c8ff213d39a7f34a36bf26db4"
        },
        "conversionRecipe": "mlx-community 4-bit (mlx_lm.convert -q 4)"
      },
      "quantization": "q4-mlx",
      "tensorParallel": 1,
      "requiredMemoryBytes": 12884901888,
      "hardware": {
        "chipFamilies": [
          "M2 Pro",
          "M2 Max",
          "M2 Ultra",
          "M3 Pro",
          "M3 Max",
          "M4",
          "M4 Pro",
          "M4 Max"
        ],
        "minUnifiedMemoryBytes": 25769803776
      },
      "perf": {
        "minThroughputTokS": 20,
        "maxConcurrencyAssumption": 1
      },
      "capabilities": {
        "tools": false,
        "structuredOutputs": false,
        "reasoning": false
      }
    }
  ]
}