Gemma 4 20B A4B IT REAP benchmark on an M4 Pro · 64 GB

{
  "runId": "run_b1a7e3ac-44ac-4018-98d2-10044747beb1",
  "bundleId": "mlx-gemma-4-21b-reap-tool-calling-mlx-4bit-e9cf2a",
  "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 20B A4B IT REAP",
    "format": "mlx",
    "quant": "4bit",
    "architecture": "gemma4",
    "source": "deadbydawn101/gemma-4-21b-REAP-Tool-Calling-mlx-4bit",
    "fileSizeBytes": 12825553410,
    "lab": {
      "name": "0xSero",
      "slug": "0xsero"
    },
    "quantizedBy": {
      "name": "Gabe Garcia",
      "slug": "deadbydawn101"
    }
  },
  "device": {
    "cpu": "Apple M4 Pro",
    "cpuCores": 14,
    "gpu": "Apple M4 Pro",
    "gpuCores": 20,
    "gpuCount": 1,
    "ramGb": 64,
    "osName": "macOS",
    "osVersion": "26.3"
  },
  "decodeTpsMean": 68.6,
  "prefillTpsMean": 754.3,
  "ttftP50Ms": 5328.72,
  "idleTpsMean": 12288,
  "peakRssMb": 14336,
  "trialsPassed": 10,
  "trialsTotal": 10,
  "runnabilityScore": 0.73458,
  "bundleSha256": "ca7421d446266d25b2398995fc5546c21f8458dd5d67ff27b55b4404724b148f",
  "createdAt": "2026-04-16T03:25:27.533Z"
}