Gemma 4 E2B benchmark on an M1 Ultra · 64 GB

{
  "runId": "run_a2e24a06-ffa2-4e5a-b25b-6290404e845b",
  "bundleId": "mlx-gemma-4-e2b-4bit-60f29c",
  "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 E2B",
    "format": "mlx",
    "quant": "4bit",
    "architecture": "gemma4",
    "source": "mlx-community/gemma-4-e2b-4bit",
    "fileSizeBytes": 3581101896,
    "lab": {
      "name": "Google",
      "slug": "google"
    },
    "quantizedBy": {
      "name": "MLX Community",
      "slug": "mlx-community"
    }
  },
  "device": {
    "cpu": "Apple M1 Ultra",
    "cpuCores": 20,
    "gpu": "Apple M1 Ultra",
    "gpuCores": 48,
    "gpuCount": 1,
    "ramGb": 64,
    "osName": "macOS",
    "osVersion": "26.4.1"
  },
  "decodeTpsMean": 102.9,
  "prefillTpsMean": 4655.2,
  "ttftP50Ms": 879.65,
  "idleTpsMean": 1852,
  "peakRssMb": 4340,
  "trialsPassed": 10,
  "trialsTotal": 10,
  "runnabilityScore": 0.97161865234375,
  "bundleSha256": "985f1e6caaa18e201a840aaad1605e2eb0e6b62eca930f779152c0e4139f67cf",
  "createdAt": "2026-04-14T03:14:04.874Z"
}