Gemma 4 E4B IT benchmark on a GeForce RTX 2080 Ti

{
  "runId": "run_630cf218-3538-4119-bea8-0681e53d1492",
  "bundleId": "llamacpp-gemma-4-e4b-it-q4_0.gguf-2b3be4",
  "status": "verified",
  "promptTokens": 40960,
  "completionTokens": 10240,
  "contextLength": 5120,
  "harness": {
    "version": "0.1.21",
    "gitSha": "unknown"
  },
  "runtime": {
    "name": "llama.cpp",
    "version": "b9110",
    "buildFlags": "metal"
  },
  "model": {
    "displayName": "Gemma 4 E4B IT",
    "format": "gguf",
    "quant": "q4_0",
    "architecture": "gemma4",
    "source": "unsloth/gemma-4-E4B-it-GGUF:gemma-4-E4B-it-Q4_0.gguf",
    "fileSizeBytes": 4836000928,
    "lab": {
      "name": "Google",
      "slug": "google"
    },
    "quantizedBy": {
      "name": "Unsloth",
      "slug": "unsloth"
    }
  },
  "device": {
    "cpu": "AMD Ryzen 7 2700X Eight-Core Processor",
    "cpuCores": 16,
    "gpu": "NVIDIA GeForce RTX 2080 Ti",
    "gpuCores": 0,
    "gpuCount": 1,
    "ramGb": 31,
    "osName": "Pop!_OS 22.04 LTS",
    "osVersion": "22.04"
  },
  "decodeTpsMean": 93.3,
  "prefillTpsMean": 3378,
  "ttftP50Ms": 1213.72,
  "idleTpsMean": 2959.3,
  "peakRssMb": 5203.4,
  "trialsPassed": 10,
  "trialsTotal": 10,
  "runnabilityScore": 0.9041496039746544,
  "bundleSha256": "80415ebaa94ed8f5d61ba24763b2b36b24b22b9f90de51460db2ed6fe0c5ab7b",
  "createdAt": "2026-05-13T15:05:56.453Z"
}