Gemma 4 E2B IT benchmark on an M3 Pro · 36 GB

{
  "runId": "run_d073414f-f0f9-4a9a-b6ec-da289cd3d43c",
  "bundleId": "llamacpp-gemma-4-e2b-it-q4_k_m.gguf-82bd8a",
  "status": "pending",
  "promptTokens": 40960,
  "completionTokens": 10240,
  "contextLength": 5120,
  "harness": {
    "version": "0.1.21",
    "gitSha": "unknown"
  },
  "runtime": {
    "name": "llama.cpp",
    "version": "b9430",
    "buildFlags": "metal"
  },
  "model": {
    "displayName": "Gemma 4 E2B IT",
    "format": "gguf",
    "quant": "q4_k_m",
    "architecture": "gemma4",
    "source": "unsloth/gemma-4-E2B-it-GGUF:gemma-4-E2B-it-Q4_K_M.gguf",
    "fileSizeBytes": 3106736256,
    "lab": {
      "name": "Google",
      "slug": "google"
    },
    "quantizedBy": {
      "name": "Unsloth",
      "slug": "unsloth"
    }
  },
  "device": {
    "cpu": "Apple M3 Pro",
    "cpuCores": 12,
    "gpu": "Apple M3 Pro",
    "gpuCores": 18,
    "gpuCount": 1,
    "ramGb": 36,
    "osName": "macOS",
    "osVersion": "26.5"
  },
  "decodeTpsMean": 62.5,
  "prefillTpsMean": 1071.9,
  "ttftP50Ms": 3820.15,
  "idleTpsMean": 2382.3,
  "peakRssMb": 4044.2,
  "trialsPassed": 10,
  "trialsTotal": 10,
  "runnabilityScore": 0.800328165922619,
  "bundleSha256": "28b670fc1042cfef72f79dcee6242f4f8bece1934ce14686eb325a5899d543f1",
  "createdAt": "2026-06-03T03:52:46.734Z"
}