LFM2.5-VL-450M benchmark on an M5 Max · 64 GB

{
  "runId": "run_22f1f18a-6064-4baa-9f9a-df9c5701bed6",
  "bundleId": "llamacpp-lfm2.5-vl-450m-f16.gguf-472b45",
  "status": "verified",
  "promptTokens": 40960,
  "completionTokens": 10240,
  "contextLength": 5120,
  "harness": {
    "version": "0.1.21",
    "gitSha": "unknown"
  },
  "runtime": {
    "name": "llama.cpp",
    "version": "b8680",
    "buildFlags": "metal"
  },
  "model": {
    "displayName": "LFM2.5-VL-450M",
    "format": "gguf",
    "quant": "f16",
    "architecture": "lfm2",
    "source": "LiquidAI/LFM2.5-VL-450M-GGUF:LFM2.5-VL-450M-F16.gguf",
    "fileSizeBytes": 711486624,
    "lab": {
      "name": "Liquid AI",
      "slug": "liquid"
    },
    "quantizedBy": {
      "name": "Liquid AI",
      "slug": "liquid"
    }
  },
  "device": {
    "cpu": "Apple M5 Max",
    "cpuCores": 18,
    "gpu": "Apple M5 Max",
    "gpuCores": 40,
    "gpuCount": 1,
    "ramGb": 64,
    "osName": "macOS",
    "osVersion": "26.4.1"
  },
  "decodeTpsMean": 469,
  "prefillTpsMean": 32426.3,
  "ttftP50Ms": 126.27,
  "idleTpsMean": 0.3,
  "peakRssMb": 1199.5,
  "trialsPassed": 10,
  "trialsTotal": 10,
  "runnabilityScore": 0.9921558925083704,
  "bundleSha256": "7668f76c9e31670fb5457d14588c9de723eb26f856519a59e9d7d909b5fbea37",
  "createdAt": "2026-04-18T00:23:09.548Z"
}