Qwen3.5-9B benchmark on an M4 Pro · 24 GB

{
  "runId": "run_4836f2d5-42e6-4b3e-91ff-80211b4d4069",
  "bundleId": "mlx-qwen3.5-9b-mlx-4bit-25156e",
  "status": "pending",
  "promptTokens": 40960,
  "completionTokens": 10240,
  "contextLength": 5120,
  "harness": {
    "version": "0.1.21",
    "gitSha": "31c7541"
  },
  "runtime": {
    "name": "mlx_lm",
    "version": "0.31.3",
    "buildFlags": null
  },
  "model": {
    "displayName": "Qwen3.5-9B",
    "format": "mlx",
    "quant": "4bit",
    "architecture": "qwen3_5",
    "source": "mlx-community/Qwen3.5-9b-MLX-4bit",
    "fileSizeBytes": 5950221072,
    "lab": {
      "name": "Qwen",
      "slug": "qwen"
    },
    "quantizedBy": {
      "name": "MLX Community",
      "slug": "mlx-community"
    }
  },
  "device": {
    "cpu": "Apple M4 Pro",
    "cpuCores": 14,
    "gpu": "Apple M4 Pro",
    "gpuCores": 20,
    "gpuCount": 1,
    "ramGb": 24,
    "osName": "macOS",
    "osVersion": "26.3.1"
  },
  "decodeTpsMean": 48.3,
  "prefillTpsMean": 463.2,
  "ttftP50Ms": 8848.16,
  "idleTpsMean": 2964,
  "peakRssMb": 7306,
  "trialsPassed": 10,
  "trialsTotal": 10,
  "runnabilityScore": 0.6389101376488094,
  "bundleSha256": "9fd4e84c9360c29df5fab55a8a5686ab349fb091731b5b43f519ff5fce4033dc",
  "createdAt": "2026-10-02T05:37:50.176Z"
}