Qwen3-Coder-30B-A3B-Instruct benchmark on an M5 Max · 128 GB

{
  "runId": "run_359fa612-6209-46a7-9809-aa2ff58b0f76",
  "bundleId": "mlx-qwen3-coder-30b-a3b-instruct-4bit-3afa67",
  "status": "verified",
  "promptTokens": 40960,
  "completionTokens": 10240,
  "contextLength": 5120,
  "harness": {
    "version": "0.1.21",
    "gitSha": "unknown"
  },
  "runtime": {
    "name": "mlx_lm",
    "version": "0.31.2",
    "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 M5 Max",
    "cpuCores": 18,
    "gpu": "Apple M5 Max",
    "gpuCores": 40,
    "gpuCount": 1,
    "ramGb": 128,
    "osName": "macOS",
    "osVersion": "26.4.1"
  },
  "decodeTpsMean": 119.7,
  "prefillTpsMean": 4498.7,
  "ttftP50Ms": 909.58,
  "idleTpsMean": 3927,
  "peakRssMb": 20480,
  "trialsPassed": 10,
  "trialsTotal": 10,
  "runnabilityScore": 0.9330357142857142,
  "bundleSha256": "02595284e166c398cae17e3b453f50881393d057af01cd477818bc65c19a9bd9",
  "createdAt": "2026-04-14T23:28:59.499Z"
}