Gemma 4 26B A4B IT (Heretic) benchmark on an M4 Pro · 64 GB

{
  "runId": "run_bb1f3807-4121-4436-9fd5-96e11b2724a2",
  "bundleId": "mlx-gemma-4-26b-a4b-it-heretic-msq-2.6bit-1b74ca",
  "status": "verified",
  "promptTokens": 1536,
  "completionTokens": 768,
  "contextLength": 768,
  "harness": {
    "version": "0.1.20",
    "gitSha": "unknown"
  },
  "runtime": {
    "name": "mlx_lm",
    "version": "0.31.2",
    "buildFlags": null
  },
  "model": {
    "displayName": "Gemma 4 26B A4B IT (Heretic)",
    "format": "mlx",
    "quant": "msq-2.6bpw",
    "architecture": "gemma4",
    "source": "mlx-community/gemma-4-26B-A4B-it-heretic-msq-2.6bit",
    "fileSizeBytes": 10145182564,
    "lab": {
      "name": "Ashar",
      "slug": "coder3101"
    },
    "quantizedBy": {
      "name": "MLX Community",
      "slug": "mlx-community"
    }
  },
  "device": {
    "cpu": "Apple M4 Pro",
    "cpuCores": 14,
    "gpu": "Apple M4 Pro",
    "gpuCores": 20,
    "gpuCount": 1,
    "ramGb": 64,
    "osName": "macOS",
    "osVersion": "26.3"
  },
  "decodeTpsMean": 64.4,
  "prefillTpsMean": 753.3,
  "ttftP50Ms": 675.13,
  "idleTpsMean": 2715,
  "peakRssMb": 11264,
  "trialsPassed": 3,
  "trialsTotal": 3,
  "runnabilityScore": 0.7482692857142856,
  "bundleSha256": "b2b34d5a6e1510de49c220378b37f2637ccca41af034958f2b3204dae8adf051",
  "createdAt": "2026-04-14T15:34:56.602Z"
}