Qwen3-Coder-30B-A3B-Instruct benchmark on an M4 Pro · 48 GB

{
  "runId": "run_9bca6705-12c2-4c9f-a223-719c8f48d46f",
  "bundleId": "mlx-qwen3-coder-30b-a3b-instruct-4bit-b2bf06",
  "status": "verified",
  "promptTokens": 40960,
  "completionTokens": 10240,
  "contextLength": 5120,
  "harness": {
    "version": "0.1.21",
    "gitSha": "unknown"
  },
  "runtime": {
    "name": "mlx_lm",
    "version": "0.31.3",
    "buildFlags": null
  },
  "model": {
    "displayName": "Qwen3-Coder-30B-A3B-Instruct",
    "format": "mlx",
    "quant": "4bit",
    "architecture": "qwen3_moe",
    "source": "mlx-community/Qwen3-Coder-30B-A3B-Instruct-4bit",
    "fileSizeBytes": 17181071994,
    "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": 48,
    "osName": "macOS",
    "osVersion": "26.5.2"
  },
  "decodeTpsMean": 64.5,
  "prefillTpsMean": 776,
  "ttftP50Ms": 5237.82,
  "idleTpsMean": 4707,
  "peakRssMb": 20480,
  "trialsPassed": 10,
  "trialsTotal": 10,
  "runnabilityScore": 0.6457785714285715,
  "bundleSha256": "7dd3c655d63ccb7a450db7bc111156606bc20e4d4cc20af926ee19b26425719d",
  "createdAt": "2026-07-03T05:16:47.993Z"
}