Gemma 4 31B IT benchmark on an M5 Pro · 48 GB

{
  "runId": "run_4fdf47d7-fda4-4859-9eb2-d00020da9741",
  "bundleId": "mlx-gemma-4-31b-it-4bit-8fbcf0",
  "status": "verified",
  "promptTokens": 40960,
  "completionTokens": 10240,
  "contextLength": 5120,
  "harness": {
    "version": "0.1.20",
    "gitSha": "unknown"
  },
  "runtime": {
    "name": "mlx_lm",
    "version": "0.31.2",
    "buildFlags": null
  },
  "model": {
    "displayName": "Gemma 4 31B IT",
    "format": "mlx",
    "quant": "4bit",
    "architecture": "gemma4",
    "source": "mlx-community/gemma-4-31b-it-4bit",
    "fileSizeBytes": 18412016832,
    "lab": {
      "name": "Google",
      "slug": "google"
    },
    "quantizedBy": {
      "name": "MLX Community",
      "slug": "mlx-community"
    }
  },
  "device": {
    "cpu": "Apple M5 Pro",
    "cpuCores": 18,
    "gpu": "Apple M5 Pro",
    "gpuCores": 20,
    "gpuCount": 1,
    "ramGb": 48,
    "osName": "macOS",
    "osVersion": "26.4.1"
  },
  "decodeTpsMean": 12.9,
  "prefillTpsMean": 351.7,
  "ttftP50Ms": 11742.08,
  "idleTpsMean": 5385,
  "peakRssMb": 23552,
  "trialsPassed": 10,
  "trialsTotal": 10,
  "runnabilityScore": 0.3760261904761904,
  "bundleSha256": "e6f0be73a1e5938b6acc9807c98a57fbf801973149a0eab29d04f01838a0a0f4",
  "createdAt": "2026-04-13T11:23:00.551Z"
}