Gemma 4 E4B benchmark on an M3 · 16 GB

{
  "runId": "run_4648206e-f5b0-4ce3-af6d-a503eb11dd5b",
  "bundleId": "mlx-gemma-4-e4b-4bit-03ec56",
  "status": "verified",
  "promptTokens": 40960,
  "completionTokens": 10240,
  "contextLength": 5120,
  "harness": {
    "version": "0.1.21",
    "gitSha": "unknown"
  },
  "runtime": {
    "name": "mlx_lm",
    "version": "0.31.2",
    "buildFlags": null
  },
  "model": {
    "displayName": "Gemma 4 E4B",
    "format": "mlx",
    "quant": "4bit",
    "architecture": "gemma4",
    "source": "mlx-community/gemma-4-e4b-4bit",
    "fileSizeBytes": 5217361182,
    "lab": {
      "name": "Google",
      "slug": "google"
    },
    "quantizedBy": {
      "name": "MLX Community",
      "slug": "mlx-community"
    }
  },
  "device": {
    "cpu": "Apple M3",
    "cpuCores": 8,
    "gpu": "Apple M3",
    "gpuCores": 10,
    "gpuCount": 1,
    "ramGb": 16,
    "osName": "macOS",
    "osVersion": "26.4"
  },
  "decodeTpsMean": 24.1,
  "prefillTpsMean": 422,
  "ttftP50Ms": 10146.9,
  "idleTpsMean": 671,
  "peakRssMb": 5699,
  "trialsPassed": 10,
  "trialsTotal": 10,
  "runnabilityScore": 0.5263759905133929,
  "bundleSha256": "6760ce04bcc6995c488777355d51411420c3a174c1a086afaf963395c28d6c09",
  "createdAt": "2026-04-24T21:12:52.819Z"
}