gpt-oss-20b benchmark on an M3 Max · 128 GB

{
  "runId": "run_4290b57f-ae83-4140-b2d1-532eadf2c3fe",
  "bundleId": "mlx-gpt-oss-20b-mxfp4-q4-9ce8f0",
  "status": "verified",
  "promptTokens": 40960,
  "completionTokens": 10240,
  "contextLength": 5120,
  "harness": {
    "version": "0.1.21",
    "gitSha": "unknown"
  },
  "runtime": {
    "name": "mlx_lm",
    "version": "0.30.2",
    "buildFlags": null
  },
  "model": {
    "displayName": "gpt-oss-20b",
    "format": "mlx",
    "quant": "mxfp4",
    "architecture": "gpt_oss",
    "source": "mlx-community/gpt-oss-20b-MXFP4-Q4",
    "fileSizeBytes": 11178569159,
    "lab": {
      "name": "OpenAI",
      "slug": "openai"
    },
    "quantizedBy": {
      "name": "MLX Community",
      "slug": "mlx-community"
    }
  },
  "device": {
    "cpu": "Apple M3 Max",
    "cpuCores": 16,
    "gpu": "Apple M3 Max",
    "gpuCores": 40,
    "gpuCount": 1,
    "ramGb": 128,
    "osName": "macOS",
    "osVersion": "26.3.1"
  },
  "decodeTpsMean": 25.9,
  "prefillTpsMean": 740.5,
  "ttftP50Ms": 5228.2,
  "idleTpsMean": 1305,
  "peakRssMb": 12288,
  "trialsPassed": 10,
  "trialsTotal": 10,
  "runnabilityScore": 0.6804214285714285,
  "bundleSha256": "6d795197f790de1d8972e1e637b9cc80506e9aa75479c40fca13fe853a22fd9e",
  "createdAt": "2026-04-16T04:08:11.843Z"
}