gpt-oss-20b benchmark on an M4 Pro · 24 GB

{
  "runId": "run_cf0d023d-102a-48df-a925-6fd641a257ae",
  "bundleId": "llamacpp-gpt-oss-20b-q5_k_m.gguf-ceae69",
  "status": "verified",
  "promptTokens": 40960,
  "completionTokens": 10240,
  "contextLength": 5120,
  "harness": {
    "version": "0.1.21",
    "gitSha": "unknown"
  },
  "runtime": {
    "name": "llama.cpp",
    "version": "0.3.0",
    "buildFlags": null
  },
  "model": {
    "displayName": "gpt-oss-20b",
    "format": "gguf",
    "quant": "q5_k_m",
    "architecture": "gpt-oss",
    "source": "unsloth/gpt-oss-20b-GGUF:gpt-oss-20b-Q5_K_M.gguf",
    "fileSizeBytes": 11717357248,
    "lab": {
      "name": "OpenAI",
      "slug": "openai"
    },
    "quantizedBy": {
      "name": "Unsloth",
      "slug": "unsloth"
    }
  },
  "device": {
    "cpu": "Apple M4 Pro",
    "cpuCores": 12,
    "gpu": "Apple M4 Pro",
    "gpuCores": 16,
    "gpuCount": 1,
    "ramGb": 24,
    "osName": "macOS",
    "osVersion": "15.7.2"
  },
  "decodeTpsMean": 64.9,
  "prefillTpsMean": 635.2,
  "ttftP50Ms": 6448.42,
  "idleTpsMean": 5220.3,
  "peakRssMb": 6357.8,
  "trialsPassed": 10,
  "trialsTotal": 10,
  "runnabilityScore": 0.6999987667410714,
  "bundleSha256": "89cbea46623a7a0abbe5d23661ed853bdc37d43af612cba1e9db6a8b9957ba4e",
  "createdAt": "2026-09-01T12:27:50.215Z"
}