Llama 3.2 3B Instruct benchmark on an M4 · 32 GB

{
  "runId": "run_aae58091-d242-4998-8432-2f3774a663ca",
  "bundleId": "mlx-llama-3.2-3b-instruct-4bit-c1054c",
  "status": "verified",
  "promptTokens": 40960,
  "completionTokens": 10240,
  "contextLength": 5120,
  "harness": {
    "version": "0.1.21",
    "gitSha": "f68adca"
  },
  "runtime": {
    "name": "mlx_lm",
    "version": "0.31.2",
    "buildFlags": null
  },
  "model": {
    "displayName": "Llama 3.2 3B Instruct",
    "format": "mlx",
    "quant": "4bit",
    "architecture": "llama",
    "source": "mlx-community/Llama-3.2-3B-Instruct-4bit",
    "fileSizeBytes": 1807496278,
    "lab": {
      "name": "Meta",
      "slug": "meta"
    },
    "quantizedBy": {
      "name": "MLX Community",
      "slug": "mlx-community"
    }
  },
  "device": {
    "cpu": "Apple M4",
    "cpuCores": 10,
    "gpu": "Apple M4",
    "gpuCores": 10,
    "gpuCount": 1,
    "ramGb": 32,
    "osName": "macOS",
    "osVersion": "15.7.4"
  },
  "decodeTpsMean": 30.8,
  "prefillTpsMean": 305.3,
  "ttftP50Ms": 13546.81,
  "idleTpsMean": 2585,
  "peakRssMb": 3465,
  "trialsPassed": 10,
  "trialsTotal": 10,
  "runnabilityScore": 0.640831396484375,
  "bundleSha256": "798e105956d554e0c5ca9207df2eae42d5fce300e69c8bc9cc309ba863f6a026",
  "createdAt": "2026-04-17T13:06:55.764Z"
}