{
  "row": 37,
  "utc": "2026-11-09T00:43:51.480088+00:00",
  "image": "sha256:45decd88eb7c10c0408327438e07c2a655e45cc7534f8b662e5c4089a6b88568",
  "container": "8d066d8a0cd98c7de8366ff6c8b7b5969c49ed3e3efba0b09cbcb9e7feabd29e",
  "runner_exit": 0,
  "manifest_sha256": "ae0dba47e379e10ae120c82dbe767b915856ebc0a56771c0751407ceb9b17f98",
  "verified_regular_files": 173,
  "closure": {
    "Status": "exited",
    "Running": false,
    "Paused": false,
    "Restarting": false,
    "OOMKilled": false,
    "Dead": false,
    "Pid": 0,
    "ExitCode": 137,
    "Error": "",
    "StartedAt": "2026-11-09T00:22:27.410864084Z",
    "FinishedAt": "2026-11-09T00:43:40.972906165Z"
  },
  "restart_count": 0,
  "complete": true,
  "strict_pass": true,
  "limits": [
    "One fresh measurement per distinct runtime variant; grouped historical rows are not independent new samples.",
    "Historical published measurements remain separate from this fresh run.",
    "Installed source and mapped native identities establish admitted runtime, not a dynamic per-kernel invocation trace.",
    "Fresh root-SSD caches and the current host state differ from the historical runs.",
    "Physical VRAM is sampled, not a continuous peak; logged allocation budgets are separate from tensor-byte totals."
  ],
  "runtime-ready": {
    "sha256": "6194544b2f07e31ce61ac20e5ffbb29ab8dee0f2c16809a798f12f391a2c35e4",
    "installed_files": 61,
    "CPU_probe_contract_sha256": "8c7ffd8cc2bfbae2ae00816e1084ef125c5f69f48f5ff4bdaddce875f10c3d6f",
    "actual_worker_PIDs_by_TP_rank": {
      "0": 227,
      "1": 228,
      "2": 229,
      "3": 230,
      "4": 231,
      "5": 232,
      "6": 233,
      "7": 234
    }
  },
  "runtime-after-ladder": {
    "sha256": "c02e8f0079c837565b928e1ca955c31c883635dea7e5e7310dd139caf2420a39",
    "installed_files": 61,
    "CPU_probe_contract_sha256": "8c7ffd8cc2bfbae2ae00816e1084ef125c5f69f48f5ff4bdaddce875f10c3d6f",
    "actual_worker_PIDs_by_TP_rank": {
      "0": 227,
      "1": 228,
      "2": 229,
      "3": 230,
      "4": 231,
      "5": 232,
      "6": 233,
      "7": 234
    }
  },
  "physical_memory": {
    "scope": "Physical device VRAM sampled every two seconds; may include allocations outside the test container. Sampled maximum is not an exact peak and not logical tensor bytes.",
    "samples": 604,
    "first_utc": "2026-11-09T00:22:27.584616+00:00",
    "last_utc": "2026-11-09T00:42:38.002078+00:00",
    "devices": [
      {
        "render": 128,
        "pci": "0000:06:00.0",
        "sampled_max_vram_bytes": 33744068608,
        "minimum_vram_bytes": 60289024
      },
      {
        "render": 129,
        "pci": "0000:09:00.0",
        "sampled_max_vram_bytes": 33634144256,
        "minimum_vram_bytes": 10858496
      },
      {
        "render": 130,
        "pci": "0000:45:00.0",
        "sampled_max_vram_bytes": 33684496384,
        "minimum_vram_bytes": 10858496
      },
      {
        "render": 131,
        "pci": "0000:48:00.0",
        "sampled_max_vram_bytes": 33634095104,
        "minimum_vram_bytes": 10862592
      },
      {
        "render": 132,
        "pci": "0000:89:00.0",
        "sampled_max_vram_bytes": 33684230144,
        "minimum_vram_bytes": 10862592
      },
      {
        "render": 133,
        "pci": "0000:8c:00.0",
        "sampled_max_vram_bytes": 33633964032,
        "minimum_vram_bytes": 10858496
      },
      {
        "render": 134,
        "pci": "0000:c5:00.0",
        "sampled_max_vram_bytes": 33684738048,
        "minimum_vram_bytes": 10854400
      },
      {
        "render": 135,
        "pci": "0000:c8:00.0",
        "sampled_max_vram_bytes": 33634250752,
        "minimum_vram_bytes": 10862592
      }
    ]
  },
  "runtime_configuration_source": {
    "container_inspect_sha256": "89f1d675db7f2a7533942ff909d5bab105018d5bed341580976f080362d19fbf",
    "preflight_sha256": "fe89ac0e5e679ad1d16154791fa5da8f7ad86831f59de1991c1dc7cb458e8056",
    "docker_run_sha256": "23efee1d0b5743ea241d8efb0a6662a68ca99f4f272082c4d53eea1e9411bf1f",
    "effective_engine_log_lines": [
      {
        "line": 7,
        "text": "2026-11-09T00:22:54.762826319Z (APIServer pid=145) INFO 11-09 00:22:54 [utils.py:228] non-default args: {'model_tag': '/opt/local-models/qwen36-27b-f16-gguf/Qwen3.6-27B-F16.gguf', 'enable_auto_tool_choice': True, 'tool_call_parser': 'qwen3_coder', 'host': '0.0.0.0', 'port': 8001, 'model': '/opt/local-models/qwen36-27b-f16-gguf/Qwen3.6-27B-F16.gguf', 'tokenizer': 'Qwen/Qwen3.6-27B', 'trust_remote_code': True, 'dtype': 'half', 'hf_config_path': '/opt/local-models/qwen36-27b-text-config-eosfix', 'tokenizer_revision': '6a9e13bd6fc8f0983b9b99948120bc37f49c13e9', 'max_model_len': 131072, 'served_model_name': ['Qwen3.6-27B-F16-GGUF-lmheadunquant'], 'hf_overrides': {'architectures': ['Qwen3_5ForCausalLM']}, 'generation_config': 'vllm', 'load_format': 'gguf', 'reasoning_parser': 'qwen3', 'tensor_parallel_size': 8, 'gpu_memory_utilization': 0.95, 'enable_prefix_caching': False, 'language_model_only': True, 'max_num_batched_tokens': 4, 'max_num_seqs': 2, 'async_scheduling': True, 'compilation_config': {'mode': None, 'debug_dump_path': PosixPath('/runwork/debug_dump'), 'cache_dir': '/runwork/compile_cache', 'compile_cache_save_format': 'unpacked', 'backend': 'inductor', 'custom_ops': [], 'splitting_ops': None, 'compile_mm_encoder': False, 'compile_sizes': None, 'compile_ranges_endpoints': None, 'inductor_compile_config': {'enable_auto_functionalized_v2': False, 'combo_kernels': True, 'benchmark_combo_kernel': True}, 'inductor_passes': {}, 'cudagraph_mode': None, 'cudagraph_num_of_warmups': 0, 'cudagraph_capture_sizes': None, 'cudagraph_copy_inputs': False, 'cudagraph_specialize_lora': True, 'use_inductor_graph_partition': True, 'pass_config': {}, 'max_cudagraph_capture_size': None, 'dynamic_shapes_config': {'type': <DynamicShapesType.BACKED: 'backed'>, 'evaluate_guards': False, 'assume_32_bit_indexing': False}, 'local_cache_dir': None, 'fast_moe_cold_start': None, 'static_all_moe_layers': []}, 'optimization_level': '3'}"
      },
      {
        "line": 55,
        "text": "2026-11-09T00:23:14.764946206Z (APIServer pid=145) INFO 11-09 00:23:14 [model.py:1562] Using max model len 131072"
      },
      {
        "line": 64,
        "text": "2026-11-09T00:23:36.349251210Z (EngineCore_DP0 pid=200) INFO 11-09 00:23:36 [core.py:101] Initializing a V1 LLM engine (v0.0.0+gfx906) with config: model='/opt/local-models/qwen36-27b-f16-gguf/Qwen3.6-27B-F16.gguf', speculative_config=None, tokenizer='/root/.cache/huggingface/hub/models--Qwen--Qwen3.6-27B/snapshots/6a9e13bd6fc8f0983b9b99948120bc37f49c13e9', skip_tokenizer_init=False, tokenizer_mode=auto, revision=None, tokenizer_revision=6a9e13bd6fc8f0983b9b99948120bc37f49c13e9, trust_remote_code=True, dtype=torch.float16, max_seq_len=131072, download_dir=None, load_format=gguf, tensor_parallel_size=8, pipeline_parallel_size=1, data_parallel_size=1, decode_context_parallel_size=1, dcp_comm_backend=ag_rs, disable_custom_all_reduce=True, quantization=gguf, enforce_eager=False, enable_return_routed_experts=False, kv_cache_dtype=auto, device_config=cuda, structured_outputs_config=StructuredOutputsConfig(backend='auto', disable_any_whitespace=False, disable_additional_properties=False, reasoning_parser='qwen3', reasoning_parser_plugin='', enable_in_reasoning=False), observability_config=ObservabilityConfig(show_hidden_metrics_for_version=None, otlp_traces_endpoint=None, collect_detailed_traces=None, kv_cache_metrics=False, kv_cache_metrics_sample=0.01, cudagraph_metrics=False, enable_layerwise_nvtx_tracing=False, enable_mfu_metrics=False, enable_mm_processor_stats=False, enable_logging_iteration_details=False), seed=0, served_model_name=Qwen3.6-27B-F16-GGUF-lmheadunquant, enable_prefix_caching=False, enable_chunked_prefill=True, pooler_config=None, compilation_config={'mode': <CompilationMode.VLLM_COMPILE: 3>, 'debug_dump_path': PosixPath('/runwork/debug_dump'), 'cache_dir': '/runwork/compile_cache', 'compile_cache_save_format': 'unpacked', 'backend': 'inductor', 'custom_ops': ['+sparse_attn_indexer', 'none'], 'splitting_ops': ['vllm::unified_attention', 'vllm::unified_attention_with_output', 'vllm::unified_mla_attention', 'vllm::unified_mla_attention_with_output', 'vllm::mamba_mixer2', 'vllm::mamba_mixer', 'vllm::short_conv', 'vllm::linear_attention', 'vllm::plamo2_mamba_mixer', 'vllm::gdn_attention_core', 'vllm::olmo_hybrid_gdn_full_forward', 'vllm::kda_attention', 'vllm::sparse_attn_indexer', 'vllm::rocm_aiter_sparse_attn_indexer'], 'compile_mm_encoder': False, 'compile_sizes': [], 'compile_ranges_endpoints': [4], 'inductor_compile_config': {'enable_auto_functionalized_v2': False, 'combo_kernels': True, 'benchmark_combo_kernel': True}, 'inductor_passes': {}, 'cudagraph_mode': <CUDAGraphMode.FULL_AND_PIECEWISE: (2, 1)>, 'cudagraph_num_of_warmups': 1, 'cudagraph_capture_sizes': [1, 2, 4], 'cudagraph_copy_inputs': False, 'cudagraph_specialize_lora': True, 'use_inductor_graph_partition': True, 'pass_config': {'fuse_norm_quant': False, 'fuse_act_quant': False, 'fuse_attn_quant': False, 'enable_sp': False, 'fuse_gemm_comms': False, 'fuse_allreduce_rms': False, 'fuse_act_padding': False, 'fuse_rope_kvcache': False}, 'max_cudagraph_capture_size': 4, 'dynamic_shapes_config': {'type': <DynamicShapesType.BACKED: 'backed'>, 'evaluate_guards': False, 'assume_32_bit_indexing': False}, 'local_cache_dir': None, 'fast_moe_cold_start': False, 'static_all_moe_layers': []}"
      }
    ]
  },
  "runtime_route_log_observations": {
    "scope": "Actual retained runtime messages, including shape-specific fast-path refusals and fallbacks. These are not a complete kernel execution trace or invocation count.",
    "source_log_sha256": "30eaedd514a475fc26437568d4aea32553d3039dc260c4f09d230e88a0ecd270",
    "lines": []
  },
  "prepared_plan": {
    "row": 37,
    "profile": "gguf-dense27b-tp8",
    "container_name": "historical-row37-d673e0b6ee-reproduction",
    "root": "/home/ai/localaiservers-historical-20261108-r3/benchmark-row-37",
    "models_root": "/usr/share/ollama/localaiservers-repro-models",
    "image": "sha256:45decd88eb7c10c0408327438e07c2a655e45cc7534f8b662e5c4089a6b88568",
    "docker_argv": [
      "docker",
      "run",
      "-d",
      "--name",
      "historical-row37-d673e0b6ee-reproduction",
      "--network",
      "host",
      "--ipc",
      "host",
      "--privileged",
      "--device",
      "/dev/kfd",
      "--device",
      "/dev/dri",
      "--group-add",
      "video",
      "--cap-add",
      "SYS_PTRACE",
      "--cap-add",
      "IPC_LOCK",
      "--security-opt",
      "seccomp=unconfined",
      "--security-opt",
      "label=disable",
      "--ulimit",
      "memlock=-1:-1",
      "--mount",
      "type=bind,src=/home/ai/localaiservers-historical-20261108-r3/benchmark-row-37/hf,dst=/root/.cache/huggingface",
      "--mount",
      "type=bind,src=/home/ai/localaiservers-historical-20261108-r3/benchmark-row-37/launch,dst=/runwork",
      "--mount",
      "type=bind,src=/home/ai/localaiservers-historical-20261108-r3/benchmark-row-37/cache/vllm,dst=/root/.cache/vllm",
      "--mount",
      "type=bind,src=/home/ai/localaiservers-historical-20261108-r3/benchmark-row-37/cache/triton,dst=/root/.triton",
      "--mount",
      "type=bind,src=/home/ai/localaiservers-historical-20261108-r3/benchmark-row-37/cache/inductor,dst=/tmp/torchinductor_root",
      "--mount",
      "type=bind,src=/home/ai/localaiservers-historical-20261108-r3/benchmark-row-37/overlay,dst=/opt/vllm_patch_bundle,readonly",
      "--mount",
      "type=bind,src=/usr/share/ollama/localaiservers-repro-models/qwen36-gguf/Qwen3.6-27B-F16.gguf,dst=/opt/local-models/qwen36-27b-f16-gguf/Qwen3.6-27B-F16.gguf,readonly",
      "--mount",
      "type=bind,src=/usr/share/ollama/localaiservers-repro-models/qwen36-27b-text-config-eosfix,dst=/opt/local-models/qwen36-27b-text-config-eosfix,readonly",
      "--mount",
      "type=bind,src=/usr/share/ollama/localaiservers-repro-models/qwen36-dense,dst=/root/.cache/huggingface/hub/models--Qwen--Qwen3.6-27B/snapshots/6a9e13bd6fc8f0983b9b99948120bc37f49c13e9,readonly",
      "-e",
      "HIP_VISIBLE_DEVICES=0,1,2,3,4,5,6,7",
      "-e",
      "LD_LIBRARY_PATH=/opt/gfx906:/rccl-overlay/install/lib:/usr/local/lib:/opt/rocm/lib:/opt/venv/lib/python3.12/site-packages/torch/lib",
      "-e",
      "VLLM_PATCH_BUNDLE=/opt/vllm_patch_bundle",
      "-e",
      "HF_HUB_OFFLINE=1",
      "--entrypoint",
      "/runwork/container_entrypoint.sh",
      "sha256:45decd88eb7c10c0408327438e07c2a655e45cc7534f8b662e5c4089a6b88568"
    ],
    "docker_run": "docker \\\n  run \\\n  -d \\\n  --name \\\n  historical-row37-d673e0b6ee-reproduction \\\n  --network \\\n  host \\\n  --ipc \\\n  host \\\n  --privileged \\\n  --device \\\n  /dev/kfd \\\n  --device \\\n  /dev/dri \\\n  --group-add \\\n  video \\\n  --cap-add \\\n  SYS_PTRACE \\\n  --cap-add \\\n  IPC_LOCK \\\n  --security-opt \\\n  seccomp=unconfined \\\n  --security-opt \\\n  label=disable \\\n  --ulimit \\\n  memlock=-1:-1 \\\n  --mount \\\n  type=bind,src=/home/ai/localaiservers-historical-20261108-r3/benchmark-row-37/hf,dst=/root/.cache/huggingface \\\n  --mount \\\n  type=bind,src=/home/ai/localaiservers-historical-20261108-r3/benchmark-row-37/launch,dst=/runwork \\\n  --mount \\\n  type=bind,src=/home/ai/localaiservers-historical-20261108-r3/benchmark-row-37/cache/vllm,dst=/root/.cache/vllm \\\n  --mount \\\n  type=bind,src=/home/ai/localaiservers-historical-20261108-r3/benchmark-row-37/cache/triton,dst=/root/.triton \\\n  --mount \\\n  type=bind,src=/home/ai/localaiservers-historical-20261108-r3/benchmark-row-37/cache/inductor,dst=/tmp/torchinductor_root \\\n  --mount \\\n  type=bind,src=/home/ai/localaiservers-historical-20261108-r3/benchmark-row-37/overlay,dst=/opt/vllm_patch_bundle,readonly \\\n  --mount \\\n  type=bind,src=/usr/share/ollama/localaiservers-repro-models/qwen36-gguf/Qwen3.6-27B-F16.gguf,dst=/opt/local-models/qwen36-27b-f16-gguf/Qwen3.6-27B-F16.gguf,readonly \\\n  --mount \\\n  type=bind,src=/usr/share/ollama/localaiservers-repro-models/qwen36-27b-text-config-eosfix,dst=/opt/local-models/qwen36-27b-text-config-eosfix,readonly \\\n  --mount \\\n  type=bind,src=/usr/share/ollama/localaiservers-repro-models/qwen36-dense,dst=/root/.cache/huggingface/hub/models--Qwen--Qwen3.6-27B/snapshots/6a9e13bd6fc8f0983b9b99948120bc37f49c13e9,readonly \\\n  -e \\\n  HIP_VISIBLE_DEVICES=0,1,2,3,4,5,6,7 \\\n  -e \\\n  LD_LIBRARY_PATH=/opt/gfx906:/rccl-overlay/install/lib:/usr/local/lib:/opt/rocm/lib:/opt/venv/lib/python3.12/site-packages/torch/lib \\\n  -e \\\n  VLLM_PATCH_BUNDLE=/opt/vllm_patch_bundle \\\n  -e \\\n  HF_HUB_OFFLINE=1 \\\n  --entrypoint \\\n  /runwork/container_entrypoint.sh \\\n  sha256:45decd88eb7c10c0408327438e07c2a655e45cc7534f8b662e5c4089a6b88568",
    "required_models": [
      "qwen36-dense",
      "qwen36-gguf"
    ],
    "mounts": [
      {
        "source": "/home/ai/localaiservers-historical-20261108-r3/benchmark-row-37/hf",
        "destination": "/root/.cache/huggingface",
        "read_only": false
      },
      {
        "source": "/home/ai/localaiservers-historical-20261108-r3/benchmark-row-37/launch",
        "destination": "/runwork",
        "read_only": false
      },
      {
        "source": "/home/ai/localaiservers-historical-20261108-r3/benchmark-row-37/cache/vllm",
        "destination": "/root/.cache/vllm",
        "read_only": false
      },
      {
        "source": "/home/ai/localaiservers-historical-20261108-r3/benchmark-row-37/cache/triton",
        "destination": "/root/.triton",
        "read_only": false
      },
      {
        "source": "/home/ai/localaiservers-historical-20261108-r3/benchmark-row-37/cache/inductor",
        "destination": "/tmp/torchinductor_root",
        "read_only": false
      },
      {
        "source": "/home/ai/localaiservers-historical-20261108-r3/benchmark-row-37/overlay",
        "destination": "/opt/vllm_patch_bundle",
        "read_only": true
      },
      {
        "source": "/usr/share/ollama/localaiservers-repro-models/qwen36-gguf/Qwen3.6-27B-F16.gguf",
        "destination": "/opt/local-models/qwen36-27b-f16-gguf/Qwen3.6-27B-F16.gguf",
        "read_only": true
      },
      {
        "source": "/usr/share/ollama/localaiservers-repro-models/qwen36-27b-text-config-eosfix",
        "destination": "/opt/local-models/qwen36-27b-text-config-eosfix",
        "read_only": true
      },
      {
        "source": "/usr/share/ollama/localaiservers-repro-models/qwen36-dense",
        "destination": "/root/.cache/huggingface/hub/models--Qwen--Qwen3.6-27B/snapshots/6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
        "read_only": true
      }
    ],
    "original_runtime_signature": "7a2a81acfc337c0f64ac6b8b0059b81d6bb4758c149c614ae6148daad82583cb",
    "differences_from_original": [
      "Fresh container name and empty cache/result paths.",
      "Model inputs are individually bound read-only at the original container paths; only verified inputs are exposed.",
      "HF_HUB_OFFLINE=1 prevents public repository changes. For originally unpinned GGUF tokenizers, the delivered cache reference explicitly binds revision 6a9e13bd6fc8f0983b9b99948120bc37f49c13e9. This is a new pinned reproduction, not proof of the original unrecorded tokenizer selection."
    ],
    "gpu_test_status": "pending",
    "cpu_affinity": "No Docker cpuset in original; host default affinity retained.",
    "port": 8001,
    "proxy_port": 19080,
    "input_package": "historical-reproduction-r3.tar.gz"
  },
  "actual_launch_contract": {
    "all_explicit_Docker_arguments_match_actual_inspect": true,
    "verified_bind_mounts": 9,
    "verified_explicit_environment": 4
  },
  "memory_log_observations": {
    "model_load_delta_GiB": [
      {
        "line": 4589,
        "text": "2026-11-09T00:25:47.689313425Z (Worker pid=227) (Worker_TP0 pid=227) INFO 11-09 00:25:47 [gpu_model_runner.py:4579] Model loading took 6.75 GiB memory and 105.144041 seconds",
        "values": [
          "6.75"
        ]
      }
    ],
    "available_KV_GiB": [
      {
        "line": 5172,
        "text": "2026-11-09T00:28:42.161027038Z (Worker pid=227) (Worker_TP0 pid=227) INFO 11-09 00:28:42 [gpu_worker.py:453] Available KV cache memory: 22.79 GiB",
        "values": [
          "22.79"
        ]
      }
    ],
    "grouped_GPU_KV_tokens": [
      {
        "line": 5175,
        "text": "2026-11-09T00:28:42.250244890Z (EngineCore_DP0 pid=200) INFO 11-09 00:28:42 [kv_cache_utils.py:1314] GPU KV cache size: 373,360 tokens",
        "values": [
          "373,360"
        ]
      }
    ],
    "context_capacity": [
      {
        "line": 5176,
        "text": "2026-11-09T00:28:42.250362411Z (EngineCore_DP0 pid=200) INFO 11-09 00:28:42 [kv_cache_utils.py:1319] Maximum concurrency for 131,072 tokens per request: 11.32x",
        "values": [
          "131,072",
          "11.32"
        ]
      }
    ]
  },
  "tiers": {
    "warmup1": {
      "completion_tokens": 2000,
      "decode_tps": 70.87263576495425,
      "prefill_seconds": 193.69643982313573,
      "client_total_seconds": 221.9462565947324,
      "strict_pass": null,
      "finish_reason": "length",
      "raw_counter_deltas": {
        "vllm:request_prefill_time_seconds_sum": 193.69643982313573,
        "vllm:prompt_tokens_total": 431.0,
        "vllm:request_decode_time_seconds_sum": 28.21963623072952,
        "vllm:request_prefill_time_seconds_count": 1.0,
        "vllm:request_decode_time_seconds_count": 1.0,
        "vllm:generation_tokens_total": 2000.0
      },
      "request_sha256": "fb3631f25efa0d86c0631caf9b534448d8eb1d0d295d20ccbddb57f4b2b8168a",
      "summary_sha256": "208c9d6d792212d91dcfeafcce4a338820decb917020c57d79517ae8ccfdac13"
    },
    "warmup2": {
      "completion_tokens": 2000,
      "decode_tps": 70.85221992214483,
      "prefill_seconds": 14.301281746476889,
      "client_total_seconds": 42.53821524605155,
      "strict_pass": null,
      "finish_reason": "length",
      "raw_counter_deltas": {
        "vllm:request_prefill_time_seconds_sum": 14.301281746476889,
        "vllm:prompt_tokens_total": 431.0,
        "vllm:request_decode_time_seconds_sum": 28.227767629548907,
        "vllm:request_prefill_time_seconds_count": 1.0,
        "vllm:request_decode_time_seconds_count": 1.0,
        "vllm:generation_tokens_total": 2000.0
      },
      "request_sha256": "fb3631f25efa0d86c0631caf9b534448d8eb1d0d295d20ccbddb57f4b2b8168a",
      "summary_sha256": "f91c140afca0dd26e2824b6e39d90ba668ef393fec8d7cec413585889f4f2110"
    },
    "warmup3": {
      "completion_tokens": 2000,
      "decode_tps": 70.90446144082557,
      "prefill_seconds": 13.469900828786194,
      "client_total_seconds": 41.684435499832034,
      "strict_pass": null,
      "finish_reason": "length",
      "raw_counter_deltas": {
        "vllm:request_prefill_time_seconds_sum": 13.469900828786194,
        "vllm:prompt_tokens_total": 431.0,
        "vllm:request_decode_time_seconds_sum": 28.206969764083624,
        "vllm:request_prefill_time_seconds_count": 1.0,
        "vllm:request_decode_time_seconds_count": 1.0,
        "vllm:generation_tokens_total": 2000.0
      },
      "request_sha256": "fb3631f25efa0d86c0631caf9b534448d8eb1d0d295d20ccbddb57f4b2b8168a",
      "summary_sha256": "453213c5b344d8c3f2231eea0fb3d1c747ea8560ea731ec584a065b1e106a139"
    },
    "warmup4": {
      "completion_tokens": 2000,
      "decode_tps": 70.89580375020118,
      "prefill_seconds": 14.021247262135148,
      "client_total_seconds": 42.240451119840145,
      "strict_pass": null,
      "finish_reason": "length",
      "raw_counter_deltas": {
        "vllm:request_prefill_time_seconds_sum": 14.021247262135148,
        "vllm:prompt_tokens_total": 431.0,
        "vllm:request_decode_time_seconds_sum": 28.210414357483387,
        "vllm:request_prefill_time_seconds_count": 1.0,
        "vllm:request_decode_time_seconds_count": 1.0,
        "vllm:generation_tokens_total": 2000.0
      },
      "request_sha256": "fb3631f25efa0d86c0631caf9b534448d8eb1d0d295d20ccbddb57f4b2b8168a",
      "summary_sha256": "8e7cde4b1ba417123d63f3db93c8db9efd360268bbf29c040ae2d4abb51e5617"
    },
    "warmup5": {
      "completion_tokens": 2000,
      "decode_tps": 70.89814051656259,
      "prefill_seconds": 14.183788473717868,
      "client_total_seconds": 42.4022008003667,
      "strict_pass": null,
      "finish_reason": "length",
      "raw_counter_deltas": {
        "vllm:request_prefill_time_seconds_sum": 14.183788473717868,
        "vllm:prompt_tokens_total": 431.0,
        "vllm:request_decode_time_seconds_sum": 28.20948455668986,
        "vllm:request_prefill_time_seconds_count": 1.0,
        "vllm:request_decode_time_seconds_count": 1.0,
        "vllm:generation_tokens_total": 2000.0
      },
      "request_sha256": "fb3631f25efa0d86c0631caf9b534448d8eb1d0d295d20ccbddb57f4b2b8168a",
      "summary_sha256": "195907145a56c7b33a822ae51523b9947bb1875fd1608ab549ff797a8af837a3"
    },
    "warmup6": {
      "completion_tokens": 2000,
      "decode_tps": 70.89678780243614,
      "prefill_seconds": 12.728598005138338,
      "client_total_seconds": 40.94778801035136,
      "strict_pass": null,
      "finish_reason": "length",
      "raw_counter_deltas": {
        "vllm:request_prefill_time_seconds_sum": 12.728598005138338,
        "vllm:prompt_tokens_total": 431.0,
        "vllm:request_decode_time_seconds_sum": 28.210022795014083,
        "vllm:request_prefill_time_seconds_count": 1.0,
        "vllm:request_decode_time_seconds_count": 1.0,
        "vllm:generation_tokens_total": 2000.0
      },
      "request_sha256": "fb3631f25efa0d86c0631caf9b534448d8eb1d0d295d20ccbddb57f4b2b8168a",
      "summary_sha256": "4420ff40180a911cf8ee9ee8add68fbfea1da19cb072db5ed22613d6b6a8295e"
    },
    "warmup7": {
      "completion_tokens": 2000,
      "decode_tps": 70.88383564402945,
      "prefill_seconds": 14.845447400584817,
      "client_total_seconds": 43.06912052445114,
      "strict_pass": null,
      "finish_reason": "length",
      "raw_counter_deltas": {
        "vllm:request_prefill_time_seconds_sum": 14.845447400584817,
        "vllm:prompt_tokens_total": 431.0,
        "vllm:request_decode_time_seconds_sum": 28.215177435427904,
        "vllm:request_prefill_time_seconds_count": 1.0,
        "vllm:request_decode_time_seconds_count": 1.0,
        "vllm:generation_tokens_total": 2000.0
      },
      "request_sha256": "fb3631f25efa0d86c0631caf9b534448d8eb1d0d295d20ccbddb57f4b2b8168a",
      "summary_sha256": "4946696a95fe3a7731ce7b6c2bb522a2f78e647719128b02c996330a77850523"
    },
    "warmup8": {
      "completion_tokens": 2000,
      "decode_tps": 70.87304671284167,
      "prefill_seconds": 12.540952987037599,
      "client_total_seconds": 40.769073138944805,
      "strict_pass": null,
      "finish_reason": "length",
      "raw_counter_deltas": {
        "vllm:request_prefill_time_seconds_sum": 12.540952987037599,
        "vllm:prompt_tokens_total": 431.0,
        "vllm:request_decode_time_seconds_sum": 28.219472602941096,
        "vllm:request_prefill_time_seconds_count": 1.0,
        "vllm:request_decode_time_seconds_count": 1.0,
        "vllm:generation_tokens_total": 2000.0
      },
      "request_sha256": "fb3631f25efa0d86c0631caf9b534448d8eb1d0d295d20ccbddb57f4b2b8168a",
      "summary_sha256": "f474753fece05f0b481c841cdf53c6707b4190899dca1810df6d8657464ef3c8"
    },
    "strict": {
      "completion_tokens": 4263,
      "decode_tps": 69.65298485444228,
      "prefill_seconds": 13.264510760083795,
      "client_total_seconds": 74.47932942770422,
      "strict_pass": true,
      "finish_reason": "stop",
      "raw_counter_deltas": {
        "vllm:request_prefill_time_seconds_sum": 13.264510760083795,
        "vllm:prompt_tokens_total": 431.0,
        "vllm:request_decode_time_seconds_sum": 61.20340727549046,
        "vllm:request_prefill_time_seconds_count": 1.0,
        "vllm:request_decode_time_seconds_count": 1.0,
        "vllm:generation_tokens_total": 4263.0
      },
      "request_sha256": "cee5b3e69c4536b401273fbab9093ac75f2e77c2992f89ab5e2c8f2cc20fcad8",
      "summary_sha256": "2a4ec12838f2054993cfc40a0ce46452682e1e06ac8c785eefd538bcc5266ae6"
    },
    "2K": {
      "completion_tokens": 2000,
      "decode_tps": 70.8759028403918,
      "prefill_seconds": 15.0672946870327,
      "client_total_seconds": 43.29480273462832,
      "strict_pass": null,
      "finish_reason": "length",
      "raw_counter_deltas": {
        "vllm:request_prefill_time_seconds_sum": 15.0672946870327,
        "vllm:prompt_tokens_total": 431.0,
        "vllm:request_decode_time_seconds_sum": 28.218335426412523,
        "vllm:request_prefill_time_seconds_count": 1.0,
        "vllm:request_decode_time_seconds_count": 1.0,
        "vllm:generation_tokens_total": 2000.0
      },
      "request_sha256": "fb3631f25efa0d86c0631caf9b534448d8eb1d0d295d20ccbddb57f4b2b8168a",
      "summary_sha256": "304d5e8b0f2afd6fcf019e0b454f8917c037eece212906cfcf2c1686136c1ee7"
    },
    "10K": {
      "completion_tokens": 10000,
      "decode_tps": 66.47442948531119,
      "prefill_seconds": 14.876342931762338,
      "client_total_seconds": 165.31903470959514,
      "strict_pass": null,
      "finish_reason": "length",
      "raw_counter_deltas": {
        "vllm:request_prefill_time_seconds_sum": 14.876342931762338,
        "vllm:prompt_tokens_total": 431.0,
        "vllm:request_decode_time_seconds_sum": 150.43378450069577,
        "vllm:request_prefill_time_seconds_count": 1.0,
        "vllm:request_decode_time_seconds_count": 1.0,
        "vllm:generation_tokens_total": 10000.0
      },
      "request_sha256": "0d4c6bf01c385fa782ffef87f014c7a329ca06561609e32989c089837866415f",
      "summary_sha256": "895a7e44c83b5069acf38a1a85397205ac0d8a1cefc38010e259fa29cda426f9"
    }
  },
  "auditor_replay": {
    "strict": {
      "exit": 0,
      "server_exit_matches": true,
      "stdout_sha256": "78c548d98b6a03634da274ae4d2f3a7efa9f86f2de678ee8658ec51a455f547a",
      "stderr_sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855"
    },
    "fixed": {
      "exit": 0,
      "server_exit_matches": true,
      "stdout_sha256": "bea8a0f9a79ac238d1154be2ce8cf7be747597bc63256c32f49b7450d42e31d4",
      "stderr_sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855"
    }
  },
  "eligible_historical_2K": true,
  "current_Qwen38_official_eligible": false,
  "archive_sha256": "aaee8a14e5491c23f2d558d10132ada9938ea7ef87544783e95a5032f095fba6",
  "archive_bytes": 73717085,
  "owner_retention_parent_exit": 0,
  "mirror_parent_exit": 0,
  "input_package": "historical-reproduction-r3.tar.gz",
  "input_package_sha256": "8c751c8c7607580bb58b9ed5005bf9b0cb59820bb0d90b8deff484c58c59691c",
  "image_archive": "gfx906-release-images.docker.tar.zst",
  "image_archive_sha256": "62f43b0328e4223420bf454ec769790c1d13c11c1c764071243d46f6e90b2715"
}
