{
  "schema": 1,
  "run_id": "2026-10-07-quick-card-prompts-b17c83",
  "status": "complete",
  "stop_reason": null,
  "suite": {
    "name": "embeddings",
    "tier": "quick",
    "suite_hash": "d499aab0d7091e3ba5b211e1cbaedf79ed4c9c86a7214156946d2ee4d21299ec",
    "ids": 66939
  },
  "subject": {
    "model": "embeddinggemma-2",
    "checkpoint": {
      "hf": "google/embeddinggemma-2",
      "revision": "914f7f89142e33e77833254d9c9b90c3cef7303b"
    },
    "quant": {
      "format": "q8",
      "producer": "onnx-community",
      "repo": "onnx-community/embeddinggemma-2-ONNX",
      "revision": "daa72c51243991dfcaf9f9137d2c573d8f7790c0"
    },
    "engine": {
      "name": "onnxruntime-cuda",
      "version": "ort1.29.0-cuda12"
    },
    "key": "embeddinggemma-2/hf-914f7f8/q8-onnx-community/onnxruntime-cuda-ort1.29.0-cuda12",
    "target": "recipe",
    "engine_visibility": "public",
    "verified_identity": true,
    "label": null
  },
  "mode": "native",
  "profile": "card-prompts",
  "harness": {
    "name": "bench",
    "commit": "d7cc28564153cb83219bc1ea0d3f83f7bc5360f4",
    "dirty": false,
    "suite_hash": "d499aab0d7091e3ba5b211e1cbaedf79ed4c9c86a7214156946d2ee4d21299ec",
    "scorer": {
      "name": "mteb",
      "version": "2.22.5"
    },
    "agent": null
  },
  "endpoint": "http://gpu-host-2/v1/embeddings",
  "dataset": {
    "tasks": [
      "NanoArguAnaRetrieval",
      "NanoClimateFeverRetrieval",
      "NanoDBPediaRetrieval",
      "NanoFEVERRetrieval",
      "NanoFiQA2018Retrieval",
      "NanoHotpotQARetrieval",
      "NanoMSMARCORetrieval",
      "NanoNFCorpusRetrieval",
      "NanoNQRetrieval",
      "NanoQuoraRetrieval",
      "NanoSCIDOCSRetrieval",
      "NanoSciFactRetrieval",
      "NanoTouche2020Retrieval",
      "STSBenchmark",
      "Banking77Classification"
    ],
    "manifest": ".artifacts/embeddings/manifests/70274443de023faf.jsonl",
    "manifest_sha256": "2d2ce7c3195335d3917cd3675666232777a4bdf38c110da28a72cac316470566",
    "texts": 66939,
    "prompts_sha256": "ad3b452b25873415721b33192ebdb7f1f848620a93446ad399b52a83540c1ff4"
  },
  "request_settings": {
    "batch_size": 32,
    "encoding_format": "base64",
    "order": "by text length, descending"
  },
  "dispatch": {
    "concurrency": 1,
    "requests_per_minute": null,
    "retries": 0
  },
  "budget": {
    "cap_usd": null,
    "spent_usd": 0,
    "price": null
  },
  "environment": {
    "node": "v24.21.0",
    "platform": "darwin",
    "arch": "arm64",
    "location": "America/Chicago"
  },
  "attempted": 2092,
  "completed": 2092,
  "started_at": "2026-10-07T04:25:02.306Z",
  "finished_at": "2026-10-07T04:54:03.912Z",
  "observed_models": {
    "embeddinggemma-2": 2092
  },
  "engine": {
    "backend": "onnxruntime",
    "onnx_file": "model_quantized.onnx",
    "requested_provider": "CUDAExecutionProvider",
    "active_providers": [
      "CUDAExecutionProvider",
      "CPUExecutionProvider"
    ],
    "onnxruntime": "1.29.0",
    "transformers": "5.19.0",
    "max_seq_length": 8192,
    "token_budget_per_forward": 16384,
    "python": "3.12.15",
    "platform": "Linux-7.0.0-34-generic-x86_64-with-glibc2.36",
    "machine": "x86_64",
    "numpy": "2.5.3"
  },
  "hardware": {
    "host": "gpu-host-2",
    "recipe": "embeddinggemma-2/p100/st-ort",
    "placements": [
      {
        "host": "gpu-host-2",
        "devices": [
          {
            "kind": "gpu",
            "label": "p100",
            "model": "P100",
            "vendor": "nvidia",
            "vram_gb": 16
          }
        ]
      }
    ],
    "backfilled": "2026-10-07: device facts copied from the operator's inventory after the run; the launcher did not record them yet"
  },
  "launcher": {
    "run_id": "2026-10-07T04-24-51-272Z",
    "profile": "ort-q8",
    "params": {
      "backend": "ort",
      "precision": "q8",
      "weights_repo": "onnx-community/embeddinggemma-2-ONNX",
      "weights_rev": "daa72c51243991dfcaf9f9137d2c573d8f7790c0",
      "engine_name": "onnxruntime-cuda",
      "engine_version": "ort1.29.0-cuda12",
      "producer": "onnx-community"
    },
    "source": {
      "repo": "local mirror",
      "commit": "98e84541b7452a60dcdd25f87c4c8020da9af9f3",
      "override": false,
      "mirror": "https://github.com/iammrduncan/inference-engines"
    },
    "acknowledgments": []
  },
  "reference": "results/embeddinggemma-2/hf-914f7f8/fp32-google/sentence-transformers-torch-cuda-st6.1.0-torch2.14.1-cu126/embeddings/2026-10-07-quick-card-prompts-d0777f",
  "vectors": {
    "path": ".artifacts/embeddings/vectors/embeddinggemma-2__hf-914f7f8__q8-onnx-community__onnxruntime-cuda-ort1.29.0-cuda12__embeddings__2026-10-07-quick-card-prompts-b17c83.npy",
    "sha256": "982442ba57b0eb784dc18bf786c8fee4fd5b276f32971ff2d7ba759bc8e3ad99",
    "shape": [
      66939,
      768
    ],
    "dtype": "float32",
    "uploaded": false
  },
  "edits": [
    {
      "date": "2026-10-17",
      "note": "Edited for publication: lab host names replaced with aliases; engine ports and local file paths removed. Measurements, request bodies and outputs are unchanged."
    }
  ]
}