Qwen3.5-0.8B benchmark on an M4 Pro · 48 GB

{
  "runId": "run_ab8934b4-3939-417d-9623-f678711f6296",
  "bundleId": "mlx-qwen3.5-0.8b-mlx-8bit-ea2ef9",
  "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": "Qwen3.5-0.8B",
    "format": "mlx",
    "quant": "8bit",
    "architecture": "qwen3_5",
    "source": "mlx-community/Qwen3.5-0.8B-MLX-8bit",
    "fileSizeBytes": 1001176958,
    "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": "26.5.1"
  },
  "decodeTpsMean": 195.5,
  "prefillTpsMean": 4357.1,
  "ttftP50Ms": 940.73,
  "idleTpsMean": 779,
  "peakRssMb": 2747,
  "trialsPassed": 10,
  "trialsTotal": 10,
  "runnabilityScore": 0.9760480608258928,
  "bundleSha256": "fec91e91ec2a2ab18566e1d75929e0913f7981c4fc64d399f0a82ba0f862081b",
  "createdAt": "2026-06-30T23:12:42.861Z"
}