Gemma 4 E2B benchmark on an M3 Pro · 18 GB

{
  "runId": "run_617ff7dc-5ad4-4e02-a789-b6196b52d8c4",
  "bundleId": "mlx-gemma-4-e2b-6bit-b3dd85",
  "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 E2B",
    "format": "mlx",
    "quant": "6bit",
    "architecture": "gemma4",
    "source": "mlx-community/gemma-4-e2b-6bit",
    "fileSizeBytes": 4740335854,
    "lab": {
      "name": "Google",
      "slug": "google"
    },
    "quantizedBy": {
      "name": "MLX Community",
      "slug": "mlx-community"
    }
  },
  "device": {
    "cpu": "Apple M3 Pro",
    "cpuCores": 11,
    "gpu": "Apple M3 Pro",
    "gpuCores": 14,
    "gpuCount": 1,
    "ramGb": 18,
    "osName": "macOS",
    "osVersion": "26.4.1"
  },
  "decodeTpsMean": 54.7,
  "prefillTpsMean": 2530.9,
  "ttftP50Ms": 1617.17,
  "idleTpsMean": 2634,
  "peakRssMb": 5501,
  "trialsPassed": 10,
  "trialsTotal": 10,
  "runnabilityScore": 0.7674160639880953,
  "bundleSha256": "5384d7a6309d95215e7b92ed730612df71243ece8bee62491000a817df694735",
  "createdAt": "2026-04-16T00:41:01.727Z"
}