Qwen3.5-35B-A3B benchmark on an M4 Max · 36 GB

{
  "runId": "run_17751509-f4bf-46d1-8afe-066383a5a86e",
  "bundleId": "llamacpp-qwen3.5-35b-a3b-q4_k_m.gguf-2a405a",
  "status": "verified",
  "promptTokens": 40960,
  "completionTokens": 10240,
  "contextLength": 5120,
  "harness": {
    "version": "0.1.20",
    "gitSha": "unknown"
  },
  "runtime": {
    "name": "llama.cpp",
    "version": "b8680",
    "buildFlags": "metal"
  },
  "model": {
    "displayName": "Qwen3.5-35B-A3B",
    "format": "gguf",
    "quant": "q4_k_m",
    "architecture": "qwen35moe",
    "source": "unsloth/Qwen3.5-35B-A3B-GGUF:Qwen3.5-35B-A3B-Q4_K_M.gguf",
    "fileSizeBytes": 22016023168,
    "lab": {
      "name": "Qwen",
      "slug": "qwen"
    },
    "quantizedBy": {
      "name": "Unsloth",
      "slug": "unsloth"
    }
  },
  "device": {
    "cpu": "Apple M4 Max",
    "cpuCores": 14,
    "gpu": "Apple M4 Max",
    "gpuCores": 32,
    "gpuCount": 1,
    "ramGb": 36,
    "osName": "macOS",
    "osVersion": "15.6.1"
  },
  "decodeTpsMean": 60.1,
  "prefillTpsMean": 1099.2,
  "ttftP50Ms": 3727.47,
  "idleTpsMean": 21057.9,
  "peakRssMb": 21148.7,
  "trialsPassed": 10,
  "trialsTotal": 10,
  "runnabilityScore": 0.599240626860119,
  "bundleSha256": "7ab29ee26d144cf4495107163fa04e00e572a2700f919763f69a408e5bfc663a",
  "createdAt": "2026-04-11T03:05:17.656Z"
}