artifacts/validation/space-runtime-benchmark.json

{
  "passed": true,
  "app_url": "https://build-small-hackathon-backyard-radiology-professor.hf.space",
  "runtime_status": {
    "mode": "local",
    "runtime": "llama.cpp",
    "runtime_revision": "b8772-bafae2765",
    "runtime_status": "ready",
    "models": [
      {
        "id": "default",
        "status": "unloaded",
        "architecture": {}
      },
      {
        "id": "medgemma-localizer",
        "status": "unloaded",
        "architecture": {}
      },
      {
        "id": "medgemma-professor",
        "status": "loaded",
        "architecture": {}
      }
    ],
    "model_revisions": [
      {
        "id": "medgemma-professor",
        "source": "unsloth/medgemma-27b-it-GGUF",
        "revision": "536e4d49c749bfca5aaa328bb7df3201e38e5d12",
        "quantization": "Q4_K_M"
      },
      {
        "id": "medgemma-localizer",
        "source": "unsloth/medgemma-1.5-4b-it-GGUF",
        "revision": "3855f948626b7ae42bccd082757f15078c53e758",
        "quantization": "Q4_K_M"
      }
    ],
    "required_models": [
      "medgemma-professor",
      "medgemma-localizer"
    ],
    "models_max": 1,
    "queue_depth": 0,
    "xraydar_available": true,
    "xraydar": {
      "source": "dnamodel/xraydar-cv",
      "weights_revision": "34aec5a6a8d639b4ebe717f2e5a5499b8f00c493",
      "code_revision": "f62b1a1404390bec697c6a6504f9d80193e8f93c"
    },
    "gpu": {
      "name": "NVIDIA L40S",
      "memory_used_mb": 20173,
      "memory_total_mb": 46068,
      "utilization_percent": 0
    },
    "detail": ""
  },
  "host_baseline_gpu_memory_mb": 0,
  "peak_gpu_memory_mb": 23951,
  "peak_application_gpu_memory_mb": 20203,
  "local_peak_gpu_memory_mb": 23951,
  "remote_peak_gpu_memory_mb": 20203,
  "runs": [
    {
      "index": 0,
      "temperature": "cold",
      "passed": true,
      "analysis_latency_ms": 39101,
      "chat_latency_ms": 12327,
      "image_sha256": "37cbe3f77ca8d19a7eba72f98306360deceb6f2852b5a51e1b55bf4451c577da",
      "model_runs": [
        {
          "role": "classifier",
          "model_id": "dnamodel/xraydar-cv",
          "model_source": "dnamodel/xraydar-cv",
          "model_revision": "34aec5a6a8d639b4ebe717f2e5a5499b8f00c493",
          "runtime": "PyTorch",
          "runtime_revision": "f62b1a1404390bec697c6a6504f9d80193e8f93c",
          "status": "ok",
          "latency_ms": 716,
          "time_to_first_token_ms": null,
          "tokens_per_second": null,
          "detail": ""
        },
        {
          "role": "vision",
          "model_id": "medgemma-localizer",
          "model_source": "unsloth/medgemma-1.5-4b-it-GGUF",
          "model_revision": "3855f948626b7ae42bccd082757f15078c53e758",
          "runtime": "llama.cpp",
          "runtime_revision": "b8772-bafae2765",
          "status": "ok",
          "latency_ms": 6681,
          "time_to_first_token_ms": null,
          "tokens_per_second": null,
          "detail": ""
        },
        {
          "role": "professor",
          "model_id": "medgemma-professor",
          "model_source": "unsloth/medgemma-27b-it-GGUF",
          "model_revision": "536e4d49c749bfca5aaa328bb7df3201e38e5d12",
          "runtime": "llama.cpp",
          "runtime_revision": "b8772-bafae2765",
          "status": "ok",
          "latency_ms": 20346,
          "time_to_first_token_ms": null,
          "tokens_per_second": null,
          "detail": ""
        },
        {
          "role": "professor-chat",
          "model_id": "medgemma-professor",
          "model_source": "unsloth/medgemma-27b-it-GGUF",
          "model_revision": "536e4d49c749bfca5aaa328bb7df3201e38e5d12",
          "runtime": "llama.cpp",
          "runtime_revision": "b8772-bafae2765",
          "status": "ok",
          "latency_ms": 10084,
          "time_to_first_token_ms": 1444,
          "tokens_per_second": 30.343351960210114,
          "detail": ""
        }
      ]
    },
    {
      "index": 1,
      "temperature": "warm",
      "passed": true,
      "analysis_latency_ms": 36255,
      "chat_latency_ms": 14805,
      "image_sha256": "37cbe3f77ca8d19a7eba72f98306360deceb6f2852b5a51e1b55bf4451c577da",
      "model_runs": [
        {
          "role": "classifier",
          "model_id": "dnamodel/xraydar-cv",
          "model_source": "dnamodel/xraydar-cv",
          "model_revision": "34aec5a6a8d639b4ebe717f2e5a5499b8f00c493",
          "runtime": "PyTorch",
          "runtime_revision": "f62b1a1404390bec697c6a6504f9d80193e8f93c",
          "status": "ok",
          "latency_ms": 453,
          "time_to_first_token_ms": null,
          "tokens_per_second": null,
          "detail": ""
        },
        {
          "role": "vision",
          "model_id": "medgemma-localizer",
          "model_source": "unsloth/medgemma-1.5-4b-it-GGUF",
          "model_revision": "3855f948626b7ae42bccd082757f15078c53e758",
          "runtime": "llama.cpp",
          "runtime_revision": "b8772-bafae2765",
          "status": "ok",
          "latency_ms": 6391,
          "time_to_first_token_ms": null,
          "tokens_per_second": null,
          "detail": ""
        },
        {
          "role": "professor",
          "model_id": "medgemma-professor",
          "model_source": "unsloth/medgemma-27b-it-GGUF",
          "model_revision": "536e4d49c749bfca5aaa328bb7df3201e38e5d12",
          "runtime": "llama.cpp",
          "runtime_revision": "b8772-bafae2765",
          "status": "ok",
          "latency_ms": 17765,
          "time_to_first_token_ms": null,
          "tokens_per_second": null,
          "detail": ""
        },
        {
          "role": "professor-chat",
          "model_id": "medgemma-professor",
          "model_source": "unsloth/medgemma-27b-it-GGUF",
          "model_revision": "536e4d49c749bfca5aaa328bb7df3201e38e5d12",
          "runtime": "llama.cpp",
          "runtime_revision": "b8772-bafae2765",
          "status": "ok",
          "latency_ms": 12907,
          "time_to_first_token_ms": 1926,
          "tokens_per_second": 30.05969430967879,
          "detail": ""
        }
      ]
    }
  ]
}